Building engineering environment monitoring method and system

Abstract

The invention relates to the technical field of constructional engineering environment monitoring, in particular to a constructional engineering environment monitoring method and system. The method comprises the following steps: performing feature extraction on a particulate matter real-time spectrum to obtain a particulate matter real-time feature spectrum; performing spectrum color conversion effect simulation on the particulate matter real-time characteristic spectrum, and performing scattering glare view angle analysis to obtain scattering glare view angle data; carrying out medium influence weight judgment according to the scattering glare view angle data to obtain particulate matter glare influence weight data; performing convolution calculation on the particulate matter glare influence weight data, and performing logic learning to obtain particulate matter glare view angle logic data; and carrying out automatic identification firmware design based on the particulate matter glare view angle logic data to obtain particulate matter glare view angle identification firmware. According to the invention, the construction engineering environment monitoring technology is optimized, so that the construction engineering environment monitoring technology is more perfect.

Description

Technical Field

[0001] The present invention relates to the technical field of construction engineering environment monitoring, and in particular to a construction engineering environment monitoring method and system. Background Art

[0002] Environmental monitoring of construction projects is an important part of the modern construction industry. Its background is rooted in the complexity of the interaction between engineering projects, the natural environment and human society. In the process of large-scale infrastructure construction and urbanization, the impact of construction projects on the environment has become an issue that cannot be ignored, including potential impacts on air quality, noise pollution, water resource utilization and soil ecology. At the same time, changes in the construction environment itself are also of great significance to project quality and construction safety. Environmental monitoring of construction projects aims to grasp the multi-dimensional dynamic information of the environment and construction site in real time through scientific monitoring technology and systematic methods, provide data support for construction, reduce environmental risks and improve project sustainability. Especially in the context of global climate change, green buildings and low-carbon construction have gradually become the core concepts of industry development, which puts higher requirements on environmental monitoring. However, a traditional method of environmental monitoring of construction projects has the problem of inaccurate analysis of light pollution scattering in dust or haze, and easy to cause light glare to construction workers under construction, endangering the safety of construction workers. Summary of the invention

[0003] Based on this, it is necessary to provide a construction project environment monitoring method and system to solve at least one of the above technical problems.

[0004] To achieve the above object, a construction engineering environment monitoring method is provided, the method comprising the following steps:

[0005] Step S1: deploying a multispectral imaging sensor at a construction site, and collecting real-time spectral data of dust particles to obtain a real-time spectrum of the particles; extracting features from the real-time spectrum of the particles to obtain a real-time feature spectrum of the particles;

[0006] Step S2: simulating the spectral color conversion effect of the real-time characteristic spectrum of the particulate matter to obtain spectral color conversion effect data; performing scattered glare viewing angle analysis on the spectral color conversion effect data to obtain scattered glare viewing angle data; determining the medium influence weight according to the scattered glare viewing angle data to obtain particulate matter glare influence weight data;

[0007] Step S3: performing convolution calculation on the particle glare impact weight data to obtain particle glare impact weight convolution data; performing logic learning on the particle glare impact weight convolution data and the scattered glare viewing angle data based on the strategy gradient algorithm to obtain particle glare viewing angle logic data;

[0008] Step S4: Design an automated recognition firmware based on the particle glare viewing angle logic data to obtain the particle glare viewing angle recognition firmware, and send the particle glare viewing angle recognition firmware to the cloud platform to perform construction engineering environmental monitoring.

[0009] The present invention collects spectral data of dust particles in real time by deploying multi-spectral imaging sensors. These data provide spectral information of particles at different wavelengths, allowing accurate identification of the physical properties and chemical composition of particles. The feature extraction process further extracts highly representative spectral features, which are the basis for subsequent analysis and judgment, and can help identify different types of particles, especially those that cause environmental pollution or health problems. The spectral color conversion effect simulation is performed on the real-time characteristic spectrum of the particles to simulate their interaction with different light sources in the actual environment. The spectral color conversion effect data provides the effect of particles on vision under different lighting conditions, especially their effect on scattered light, which in turn affects the visual experience of the human eye. At this time, a scattered glare viewing angle analysis is performed to help understand how particles produce glare interference to the human eye through scattered light, especially in high-risk places such as construction sites, where the glare caused by particles increases workers' visual fatigue and accident risks. Based on this analysis, the calculated scattered glare viewing angle data can further quantify and evaluate the degree of harm of particles to the environment. By performing convolution calculations on the weighted data of the particle glare impact, a more sophisticated glare impact prediction model can be obtained, and then the policy gradient algorithm can be used for logical learning. At this stage, the artificial intelligence algorithm conducts an in-depth analysis of the relationship between the scattered glare data and the particle characteristics, and optimizes the learning model to accurately identify the glare interference caused by particles under different environmental conditions. Through this process, the system can automatically adjust its adaptability and response strategy to environmental factors to improve the accuracy and real-time performance of monitoring. By combining all the above analysis results, an automatic recognition firmware design is performed based on the particle glare perspective logic data. This firmware can perform real-time monitoring in on-site equipment, identify potential hazards and issue early warnings. By connecting this recognition firmware to the cloud platform, remote data analysis and processing can be achieved to ensure more efficient environmental monitoring and safety control at the construction site. At this point, the cloud platform can not only store and share data, but also analyze historical data, optimize monitoring strategies, and improve the environmental quality and safety level of the construction site. Therefore, the present invention optimizes a traditional construction project environment monitoring method, solves the problems of inaccurate light pollution scattering analysis in dust or haze conditions, and easy light glare to construction workers during construction, endangering the safety of construction workers. It improves the accuracy of light pollution scattering analysis in dust or haze conditions, can warn construction workers of light glare caused during construction, reduces light pollution, and improves the safety of construction workers.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: deploying a multispectral imaging sensor at the construction site, and collecting real-time spectral data of dust particles to obtain a real-time spectrum of the particles;

[0012] Step S12: performing baseline correction processing on the real-time spectrum of the particulate matter to obtain a real-time baseline-corrected spectrum of the particulate matter;

[0013] Step S13: removing noise from the particle real-time baseline correction spectrum to obtain the particle real-time baseline correction denoised spectrum;

[0014] Step S14: extracting features from the real-time baseline-corrected and denoised spectrum of the particles to obtain a real-time feature spectrum of the particles.

[0015] The present invention can obtain high-precision particle spectral information by deploying multi-spectral imaging sensors at the construction site and collecting real-time spectral data of dust particles. This process can reflect the concentration and type of particles in the on-site environment in real time, and provide a reliable spectral data basis for subsequent analysis. It has strong real-time performance and ensures the timeliness and accuracy of monitoring data. Baseline correction processing is performed on the real-time spectral data of particles, which can eliminate the baseline drift caused by equipment or environmental factors and ensure the stability and consistency of data. Baseline correction can effectively reduce system errors, improve the accuracy of data, provide a clear spectral benchmark for further analysis, and help subsequent feature extraction and scientific decision-making. There is noise in the spectral data after real-time baseline correction of particles. Through noise removal processing, irrelevant signals caused by environmental interference or sensor errors can be eliminated. The spectral data after denoising is purer, and the data quality is improved. This process can effectively reduce the interference of noise on data analysis results and enhance the accuracy of particle monitoring. Feature extraction is performed on the denoised spectral data, and characteristic spectra that can represent key information such as particle type, morphology, and concentration can be extracted from it. This processing can transform the original spectral data into more informative characteristic data, making subsequent analysis and judgment more accurate. The characteristic spectrum provides a scientific basis for the classification and identification of particulate matter and further environmental impact assessment.

[0016] Preferably, step S2 comprises the following steps:

[0017] Step S21: Acquire real-time environmental humidity data and environmental light intensity change data;

[0018] Step S22: simulating the spectrum color conversion effect of the real-time characteristic spectrum of the particulate matter according to the real-time humidity data of the environment to obtain the spectrum color conversion effect data;

[0019] Step S23: performing scattered glare viewing angle analysis on the spectral color conversion effect data based on the ambient light intensity change data to obtain scattered glare viewing angle data;

[0020] Step S24: determining the medium influence weight of the real-time characteristic spectrum of the particle according to the scattered glare viewing angle data, and obtaining the particle glare influence weight data.

[0021] The present invention can provide key environmental parameters for subsequent particle spectral analysis by collecting environmental humidity and light intensity change data in real time. These data are basic information that affects the optical properties of particles such as scattering and reflection, and are of great significance for understanding the behavior of particles under different environmental conditions. Accurate environmental data can improve the accuracy and scientificity of subsequent analysis. According to the real-time humidity data, the color conversion effect of the real-time characteristic spectrum of the particles is simulated, and the spectral performance of the particles under different humidity conditions can be studied. Humidity changes will affect the hygroscopicity and reflectivity of the particles, thereby changing their spectral characteristics. By simulating these changes, the visibility and response of the particles to light under different environmental conditions can be predicted, providing a theoretical basis for monitoring and controlling the impact of particles. Based on the ambient light intensity change data, combined with the spectral color conversion effect data, the scattered glare viewing angle analysis can be performed to quantify the scattering effect of particles under different light intensities. Scattered glare can significantly affect visual clarity and safety, especially in scenes such as construction projects. Therefore, this analysis process helps to evaluate the degree of visual interference of particles and guide environmental optimization and particle control measures. By determining the medium influence weight of the characteristic spectrum of particulate matter based on the scattered glare viewing angle data, the influence of different media (such as moisture, light, temperature, etc.) on the optical properties of particulate matter can be evaluated. By determining the glare influence weight of particulate matter, its impact on the environment and human vision can be more accurately predicted, thus providing data support for optimizing air quality and visual comfort in construction projects.

[0022] Preferably, step S22 includes the following steps:

[0023] Step S221: calculating the particle size mean difference of the particle real-time characteristic spectrum to obtain the particle size mean difference;

[0024] Step S222: performing particle concentration change trend analysis on the particle real-time characteristic spectrum to obtain particle concentration change trend data;

[0025] Step S223: analyzing the particle humidity adsorption effect on the particle concentration change trend data and the particle size average difference based on the real-time humidity data of the environment to obtain the particle humidity adsorption effect data;

[0026] Step S224: performing light scattering band intensity analysis on the real-time characteristic spectrum of the particles according to the humidity adsorption effect data of the particles to obtain humidity adsorption scattering band intensity data;

[0027] Step S225: simulating the spectrum color conversion effect on the humidity adsorption scattering band intensity data to obtain spectrum color conversion effect data.

[0028] The present invention calculates the mean difference of particle size on the real-time characteristic spectrum of particles, and can accurately determine the particle size distribution characteristics of particles. Particle size is an important factor affecting the optical properties of particles. Particles with different particle sizes have different performances in light absorption, scattering, etc. By calculating the mean difference of particle size, key parameters can be provided for subsequent humidity adsorption effect analysis, thereby improving the accuracy of particle monitoring. The concentration change trend analysis of the real-time characteristic spectrum of particles can reveal the concentration fluctuation law of particles at the construction site. Understanding the change trend of particle concentration over time is helpful to predict the diffusion and accumulation of particles, and then formulate reasonable prevention and control measures. This analysis can provide important dynamic data support for environmental management and pollution control. Based on the real-time humidity data of the environment, the humidity adsorption effect analysis of the concentration change trend and particle size mean difference of particles can be carried out, and the adsorption effect of humidity on particles can be deeply studied. Humidity changes usually affect the surface properties and adsorption capacity of particles, thereby changing their optical properties. Through this analysis, the adsorption effect data of particles under different humidity conditions can be obtained, providing an important basis for subsequent optical scattering analysis. According to the data of the humidity adsorption effect of particles, the light scattering band intensity of the real-time characteristic spectrum of particles is further analyzed, which can analyze in detail the effect of humidity on the light scattering characteristics of particles. Humidity adsorption will affect the reflection and scattering ability of particles. Analyzing the changes in scattering intensity in different bands will help to accurately evaluate the impact of humidity on the visibility and environmental optical effects of particles, and provide a more accurate reference for subsequent monitoring and early warning. The spectral color conversion effect simulation of the humidity adsorption scattering band intensity data can take into account the impact of humidity on the color performance of particles and simulate the impact of humidity changes on visual effects. This simulation can predict the spectral color changes of particles under different environmental humidity, further analyze the visual impact of particles in actual environments, and provide strong support for environmental comfort assessment and particle control in construction projects.

[0029] Preferably, performing light scattering band intensity analysis on the real-time characteristic spectrum of the particles according to the humidity adsorption effect data of the particles comprises the following steps:

[0030] The particle adsorption capacity is evaluated on the particle humidity adsorption effect data to obtain the particle adsorption capacity data;

[0031] According to the particle adsorption capacity data, the spectrum width change analysis of the real-time characteristic spectrum of the particles is performed to obtain the spectrum width change data of the particles;

[0032] Performing spectral shift measurement on the particle spectral width variation data to obtain spectral shift measurement data;

[0033] The light scattering band intensity is analyzed based on the particle spectral width change data and spectral shift measurement data to obtain the humidity adsorption scattering band intensity data.

[0034] In the embodiment of the present invention, by evaluating the adsorption capacity of the particle humidity adsorption effect data, the effect of humidity on the adsorption capacity of the particle surface can be quantified. The change in adsorption capacity directly reflects the effect of humidity on the physical properties of the particle, especially the change in the surface area and surface energy of the particle. This evaluation provides a reliable basis for the subsequent light scattering analysis, and can more accurately reflect the effect of humidity on the scattering characteristics of the particle. According to the particle adsorption capacity data, the spectrum width change analysis of the real-time characteristic spectrum of the particle can reveal the effect of humidity change on the spectral characteristics of the particle (especially the spectrum width). Humidity changes usually cause changes in the light scattering characteristics of the particle, especially the increase or decrease of the spectrum width. By analyzing the change in spectrum width, the optical performance of the particle under different humidity conditions can be deeply understood, providing more refined input data for subsequent scattering intensity analysis. The spectral shift measurement of the particle spectrum width change data is intended to quantify the spectral shift of the particle under different humidity conditions. This measurement can help evaluate the changes in the optical behavior of the particle under humidity changes, such as the red shift or blue shift effect of the spectrum. By measuring the degree of spectrum shift, the effect of humidity on the scattering band of the particle can be more accurately described, providing a scientific basis for further analysis of the intensity change of the scattering band. The light scattering band intensity analysis based on the particle spectral width change data and spectral shift measurement data can accurately calculate the scattering intensity of particles in different bands under the humidity adsorption effect. Humidity-adsorbed particles will have a significant impact on the scattering of light. Through this analysis step, the light scattering characteristics of particles under different humidity conditions can be obtained, especially the scattering intensity in each band. These data provide a basis for the quantitative evaluation of the impact of particles on environmental optics and provide a basis for further visual interference control and environmental governance.

[0035] Preferably, step S23 includes the following steps:

[0036] Step S231: calculating the ultraviolet intensity period variance of the ambient light intensity change data to obtain ultraviolet intensity period variance data;

[0037] Step S232: performing ultraviolet wavelength switching frequency analysis on the ambient light intensity variation data according to the ultraviolet intensity period variance data to obtain ultraviolet wavelength switching frequency data;

[0038] Step S233: obtaining the real-time location of construction workers on the construction project;

[0039] Step S234: performing scattered glare enhancement calculation on the spectrum color conversion effect data according to the ultraviolet wavelength switching frequency data to obtain spectrum color scattered glare enhancement data;

[0040] Step S235: performing scattered glare viewing angle analysis on the real-time position of construction workers on the construction project based on the spectral color scattered glare enhancement data to obtain scattered glare viewing angle data.

[0041] The present invention can effectively evaluate the volatility and periodic change of ultraviolet intensity by calculating the periodic variance of ultraviolet intensity on the ambient light intensity change data. The fluctuation of ultraviolet intensity directly affects the scattering, absorption and impact on the construction environment of light, especially when the sunlight is strong or the shadow changes. The periodic variance data provides a basis for the subsequent ultraviolet wavelength switching frequency analysis, helps to understand the changing trend of ultraviolet intensity in different time periods, and thus provides data support for the optical control of the construction environment. According to the ultraviolet intensity periodic variance data, the ultraviolet wavelength switching frequency analysis is performed on the ambient light intensity change data, which can reveal the wavelength change frequency of ultraviolet rays under different lighting conditions. The frequency of ultraviolet wavelength change will affect the color performance and spectral distribution of light, and further affect the effect of light scattering. Through this analysis, the law of ultraviolet wavelength change can be more accurately grasped, which provides an important basis for evaluating the risk of construction personnel being exposed to ultraviolet rays, and guides the subsequent spectral color conversion effect analysis. Obtaining the real-time position of construction personnel on the construction project can provide accurate spatial positioning data for the subsequent scattered glare analysis. At the construction site, there are differences in the influence of ultraviolet intensity and wavelength on different positions. By obtaining the position of the construction workers in real time, the angle and intensity of ultraviolet radiation on the construction workers can be accurately calculated, so as to evaluate the scattering and glare effects caused by ultraviolet rays. This step helps to improve the accuracy of safety monitoring and environmental control. According to the ultraviolet wavelength switching frequency data, the scattered glare enhancement calculation of the spectral color conversion effect data can quantify the impact of ultraviolet wavelength changes on spectral colors, especially the enhancement of color changes and light scattering under high ultraviolet intensity. This calculation helps to evaluate how the scattering effect of ultraviolet wavelength switching frequency on light enhances glare and further affects the visual environment of construction workers. It can help improve the lighting design of the construction site and reduce the visual interference and safety hazards caused by glare. Based on the spectral color scattered glare enhancement data, the scattered glare perspective analysis of the real-time position of the construction workers in the construction project can help evaluate the specific impact of ultraviolet scattering on the construction workers. There are differences in ultraviolet scattering intensity and glare effect in different positions. By analyzing the perspective data of scattered glare, the visual impact of construction workers in different positions can be accurately predicted. This analysis helps to optimize the working environment of construction workers, reduce visual fatigue and safety hazards caused by excessive or weak light, and improve work efficiency and safety.

[0042] Preferably, the scattered glare enhancement calculation is performed on the spectrum color conversion effect data according to the ultraviolet wavelength switching frequency data by using a scattered glare intensity calculation formula, wherein the scattered glare intensity calculation formula is as follows:

[0043] ;

[0044] In the formula, Indicates the scattered glare intensity value, represents the mean value of the ultraviolet wavelength switching frequency, represents a natural constant, Indicates the degree of dispersion of ultraviolet wavelengths. Represents the color saturation of the spectral color conversion effect data, Indicates the wavelength range value of the ultraviolet wavelength switching frequency data, Indicates the central wavelength of the UV wavelength switching frequency data, The scattering attenuation index representing the switching frequency data of the ultraviolet wavelength, Indicates the UV dispersion coefficient in the UV wavelength switching frequency data, Represents the error correction value of the scattered glare intensity calculation formula.

[0045] The present invention provides a formula for calculating the intensity of scattered glare, which comprehensively considers multiple variables, making the calculation of scattered glare more comprehensive and able to reflect the impact of different factors on spectral color conversion. , represents the overall intensity of ultraviolet wavelengths in the spectrum. By using the mean, the influence of different wavelengths can be better balanced; the natural constant , used in the formula for exponential functions, enhances the nonlinear characteristics of the spectral response and can more accurately simulate the behavior of light in a real environment; the degree of dispersion of ultraviolet wavelengths , reflects the change of ultraviolet wavelength, affects the scattering characteristics of light, and can accurately model the response of light of different wavelengths; color saturation of spectral color conversion effect data , which affects the effect of spectral color conversion. The higher the saturation, the stronger the color perception of light, which directly affects the perception of scattered glare; the wavelength range value of ultraviolet wavelength switching frequency data , describes the distribution width of the ultraviolet wavelength switching frequency data. The larger the range, the more complex the spectral response, and can effectively simulate a variety of ambient light conditions; the central wavelength of the ultraviolet wavelength switching frequency data , which represents the main characteristic wavelength of the spectrum, has a direct impact on the scattering effect and can help determine the most sensitive wavelength region; the scattering attenuation index of the ultraviolet wavelength switching frequency data , which indicates the intensity attenuation of light due to scattering during propagation, affecting the final intensity of scattered glare; the ultraviolet dispersion coefficient in the ultraviolet wavelength switching frequency data , describes the dispersion characteristics in the ultraviolet wavelength switching frequency data, affects the width and distribution of the spectrum, and helps to simulate the characteristics of various light sources; error correction value of the scattered glare intensity calculation formula , used to correct potential errors in the formula and ensure the accuracy and reliability of the calculation results.

[0046] Preferably, performing scattered glare viewing angle analysis on the real-time position of construction workers on a construction project based on spectral color scattered glare enhancement data comprises the following steps:

[0047] Analyze the visual movement range based on the real-time position of construction workers in the construction project to obtain the visual movement range of the workers;

[0048] Calculate the scattering angle of the spectral color scattering glare enhancement data to obtain the spectral color scattering angle data;

[0049] According to the spectral color scattering angle data and the spectral color scattering glare enhancement data, the vision movement glare stimulation range of the personnel is matched to obtain the vision movement glare stimulation range;

[0050] The scattered glare visual angle analysis is performed on the visual motion glare stimulation range to obtain the scattered glare visual angle data.

[0051] The present invention analyzes the visual movement range based on the real-time position of the construction workers in the construction project, and can accurately estimate the visual activity area of ​​the construction workers during the operation. The analysis takes into account the real-time position and movement trajectory of the personnel, and helps predict the area exposed to different lighting conditions. Through this step, the risk of construction workers being exposed to potential glare sources during work can be understood, thereby providing an important reference for subsequent optical intervention and safety design. Calculating the scattering angle of the spectral color scattering glare enhancement data helps to understand the influence of ultraviolet rays and other light sources on the scattering effect at different angles. The scattering angle data can accurately describe the direction and degree of deviation of light after encountering objects (such as buildings, equipment or environmental structures) at the construction site. Through this calculation, the scattering effects of different light sources on construction workers in different positions can be quantitatively analyzed, providing physical parameter support for subsequent vision impact assessment. According to the spectral color scattering angle data and the spectral color scattering glare enhancement data, the visual movement range of the personnel is matched with the visual movement glare stimulation range. This step simulates the glare stimulation area encountered by construction workers during movement by combining the scattering angle of light with the spectral change. The matching results can help identify which areas of the construction site have strong glare stimulation, and then evaluate the degree of exposure of construction workers to glare. This step is crucial to optimizing the working environment, reducing visual fatigue and improving safety. Scattered glare perspective analysis of the range of visual movement glare stimulation helps to quantify the angular relationship between the construction workers' line of sight and the scattered light source in different working positions. This analysis can reveal the glare intensity and visual impact in different directions, thereby providing a basis for the reasonable layout of light sources and the visual comfort of construction workers. By calculating the scattered glare perspective data, the relative relationship between the light source and the construction workers' line of sight can be accurately determined, and the optical conditions in the working environment can be further optimized to reduce the discomfort and potential safety hazards caused by glare.

[0052] Preferably, step S3 comprises the following steps:

[0053] Step S31: normalizing the particle glare impact weight data to obtain particle glare impact weight normalized data;

[0054] Step S32: performing convolution calculation on the particle glare influence weight normalized data to obtain particle glare influence weight convolution data;

[0055] Step S33: Based on the policy gradient algorithm, logic learning is performed on the particle glare impact weight convolution data and the scattered glare viewing angle data to obtain the particle glare viewing angle logic data.

[0056] The present invention normalizes the particle glare impact weight data, and can convert the data into a unified dimension, so that each data item is comparable. Through normalization, the difference in different data ranges is eliminated, making the subsequent processing more stable and easy to calculate. This step is of great significance for eliminating noise and reducing the interference of outliers, and provides more standardized input data for subsequent convolution calculations and logical learning, thereby improving the accuracy and effect of subsequent analysis. Convolution calculations are performed on the normalized particle glare impact weight data, and local weighting and feature extraction of the data can be performed by applying filters or kernel functions. This process helps to extract more accurate and useful information from complex ambient lighting data, especially the scattering effect of particles on light and its impact on glare. The results of the convolution calculation can help reveal the potential patterns or correlations in the data, thereby providing more accurate input in subsequent steps and improving the understanding and application of scattered glare viewing angle data. Based on the policy gradient algorithm, logical learning is performed on the particle glare impact weight convolution data and scattered glare viewing angle data, and the decision-making process can be dynamically optimized by reinforcement learning methods. The policy gradient algorithm can learn the optimal decision-making strategy based on different input data (such as the glare effect of particulate matter and the viewing angle of scattered glare). This process can not only gradually optimize the response to glare stimuli under the influence of particulate matter, but also adjust the behavior of the system to make the optical environment more in line with safety and comfort requirements. Through logical learning, the system can automatically adapt to different environmental changes, effectively improving the adaptability and predictive ability of the model.

[0057] Preferably, the present invention further provides a construction engineering environment monitoring system for executing the construction engineering environment monitoring method as described above, the construction engineering environment monitoring system comprising:

[0058] The spectral feature extraction module is used to deploy multi-spectral imaging sensors at the construction site, collect real-time spectral data of dust particles, and obtain real-time spectra of particles; extract features from the real-time spectra of particles to obtain real-time feature spectra of particles;

[0059] The scattered glare analysis module is used to simulate the spectral color conversion effect of the real-time characteristic spectrum of the particulate matter to obtain the spectral color conversion effect data; perform scattered glare viewing angle analysis on the spectral color conversion effect data to obtain scattered glare viewing angle data; determine the medium influence weight based on the scattered glare viewing angle data to obtain the particulate matter glare influence weight data;

[0060] A logic learning module is used to perform convolution calculation on the particle glare impact weight data to obtain particle glare impact weight convolution data; based on the strategy gradient algorithm, logic learning is performed on the particle glare impact weight convolution data and the scattered glare viewing angle data to obtain particle glare viewing angle logic data;

[0061] The firmware design module is used to design automatic recognition firmware based on the particle glare perspective logic data, obtain the particle glare perspective recognition firmware, and send the particle glare perspective recognition firmware to the cloud platform to perform construction project environmental monitoring.

[0062] The beneficial effect of the present invention is that the spectral data of dust particles are collected in real time by deploying multi-spectral imaging sensors. These data provide spectral information of particles at different wavelengths, allowing accurate identification of the physical properties and chemical composition of particles. The feature extraction process further extracts highly representative spectral features, which are the basis for subsequent analysis and judgment, and can help identify different types of particles, especially those that cause environmental pollution or health problems. The spectral color conversion effect is simulated for the real-time characteristic spectrum of the particles to simulate their interaction with different light sources in the actual environment. The spectral color conversion effect data provides the effect of particles on vision under different lighting conditions, especially their effect on scattered light, which in turn affects the visual experience of the human eye. At this time, a scattered glare viewing angle analysis is performed to help understand how particles produce glare interference to the human eye through scattered light, especially in high-risk places such as construction sites, where the glare caused by particles increases workers' visual fatigue and accident risks. Based on this analysis, the calculated scattered glare viewing angle data can further quantify and evaluate the degree of harm of particles to the environment. By performing convolution calculations on the weighted data of the particle glare impact, a more sophisticated glare impact prediction model can be obtained, and then the policy gradient algorithm can be used for logical learning. At this stage, the artificial intelligence algorithm conducts an in-depth analysis of the relationship between the scattered glare data and the particle characteristics, and optimizes the learning model to accurately identify the glare interference caused by particles under different environmental conditions. Through this process, the system can automatically adjust its adaptability and response strategy to environmental factors to improve the accuracy and real-time performance of monitoring. By combining all the above analysis results, an automatic recognition firmware design is performed based on the particle glare perspective logic data. This firmware can perform real-time monitoring in on-site equipment, identify potential hazards and issue early warnings. By connecting this recognition firmware to the cloud platform, remote data analysis and processing can be achieved to ensure more efficient environmental monitoring and safety control at the construction site. At this point, the cloud platform can not only store and share data, but also analyze historical data, optimize monitoring strategies, and improve the environmental quality and safety level of the construction site. Therefore, the present invention optimizes a traditional construction project environment monitoring method, solves the problems of inaccurate light pollution scattering analysis in dust or haze conditions, and easy light glare to construction workers during construction, endangering the safety of construction workers. It improves the accuracy of light pollution scattering analysis in dust or haze conditions, can warn construction workers of light glare caused during construction, reduces light pollution, and improves the safety of construction workers. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A schematic diagram of the steps of a construction engineering environmental monitoring method;

[0064] Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG.

[0065] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0066] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0067] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0068] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0069] To achieve this, please refer to Figure 1 to Figure 2 , a construction engineering environment monitoring method, the method comprising the following steps:

[0070] Step S1: deploying a multispectral imaging sensor at a construction site, and collecting real-time spectral data of dust particles to obtain a real-time spectrum of the particles; extracting features from the real-time spectrum of the particles to obtain a real-time feature spectrum of the particles;

[0071] Step S2: simulating the spectral color conversion effect of the real-time characteristic spectrum of the particulate matter to obtain spectral color conversion effect data; performing scattered glare viewing angle analysis on the spectral color conversion effect data to obtain scattered glare viewing angle data; determining the medium influence weight according to the scattered glare viewing angle data to obtain particulate matter glare influence weight data;

[0072] Step S3: performing convolution calculation on the particle glare impact weight data to obtain particle glare impact weight convolution data; performing logic learning on the particle glare impact weight convolution data and the scattered glare viewing angle data based on the strategy gradient algorithm to obtain particle glare viewing angle logic data;

[0073] Step S4: Design an automated recognition firmware based on the particle glare viewing angle logic data to obtain the particle glare viewing angle recognition firmware, and send the particle glare viewing angle recognition firmware to the cloud platform to perform construction engineering environmental monitoring.

[0074] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a construction engineering environment monitoring method of the present invention. In this example, the construction engineering environment monitoring method includes the following steps:

[0075] Step S1: deploying a multispectral imaging sensor at a construction site, and collecting real-time spectral data of dust particles to obtain a real-time spectrum of the particles; extracting features from the real-time spectrum of the particles to obtain a real-time feature spectrum of the particles;

[0076] In an embodiment of the present invention, at a construction site, a multispectral imaging sensor is arranged at an appropriate height using a fixed mounting bracket to ensure that its field of view covers the key parts of the construction area and to avoid the equipment being blocked or contaminated. The optical components of the multispectral imaging sensor are focused and calibrated to ensure that the spectral signal is clear and reduce stray light interference. The sensor decomposes the captured light into spectral information of multiple frequency bands through a built-in multi-band filter. During the real-time spectral data acquisition process, the band scanning modules of the sensor are activated in sequence to convert the spectral information of each band from visible light to near-infrared light into electrical signals, and store them in a digital manner to finally form real-time spectral data of particulate matter. For the real-time spectral data of particulate matter, the wavelength segmentation method is used to extract the main characteristic areas, including the dispersion peak position, spectral bandwidth and peak height information. This process uses the Gaussian curve fitting method to segmentally fit the spectrum and extract local extreme points and related parameters. Subsequently, the baseline correction method is used to eliminate background light interference, and the real-time characteristic spectrum of particulate matter is obtained after removing low-frequency noise. After this step is completed, the characteristic spectrum is stored in a matrix form, and each column represents the light intensity value at a certain wavelength for subsequent analysis.

[0077] Step S2: simulating the spectral color conversion effect of the real-time characteristic spectrum of the particulate matter to obtain spectral color conversion effect data; performing scattered glare viewing angle analysis on the spectral color conversion effect data to obtain scattered glare viewing angle data; determining the medium influence weight according to the scattered glare viewing angle data to obtain particulate matter glare influence weight data;

[0078] In the embodiment of the present invention, the real-time characteristic spectrum of the particulate matter is used to simulate the spectral change effect of the particulate matter under different environmental humidity and illumination conditions according to the color perception conversion theory. The process first loads the real-time humidity data and the ambient light intensity change data, fits the humidity and spectral peak drift relationship through the Lagrange interpolation algorithm, and derives the influence of the humidity adsorption effect on the spectral change. Then, based on the light scattering principle, the weakening ratio of the spectral intensity under different illumination conditions is calculated to simulate the spectral color conversion effect. The humidity and illumination effects are superimposed by matrix multiplication to generate spectral color conversion effect data. Subsequently, the intensity distribution of scattered light at different angles is calculated for the spectral color conversion effect data using the Mie scattering formula to determine the directional characteristics of the glare effect. The scattering intensity corresponding to each viewing angle is used to generate a glare viewing angle intensity distribution diagram by matrix decomposition. Combined with these data, a weight determination rule is established, and based on the optical propagation model of discrete media, the scattered light intensity is associated with variables such as particulate matter concentration, particle size distribution, and humidity, the medium influence weight is calculated, and the particulate matter glare influence weight data is generated.

[0079] Step S3: performing convolution calculation on the particle glare impact weight data to obtain particle glare impact weight convolution data; performing logic learning on the particle glare impact weight convolution data and the scattered glare viewing angle data based on the strategy gradient algorithm to obtain particle glare viewing angle logic data;

[0080] In an embodiment of the present invention, the particle glare impact weight data is used as input, and the convolution kernel design technology is used to perform the convolution calculation of the weight data. In the convolution kernel design process, the spatial filtering parameters based on Fourier transform are set to enhance the weight characteristics of a specific frequency range. The convolution process uses the particle weight data as the input matrix, and performs weighted superposition in a two-dimensional space through a sliding window to form the particle glare impact weight convolution data. Based on the convolution data and the scattered glare viewing angle data, the policy gradient algorithm is used to construct the logic learning rules. First, the parameter matrix of the logistic regression model is initialized, and the two sets of data are mapped to the high-dimensional feature space respectively. The weight of the parameter matrix is ​​adjusted by the gradient ascent method so that the output logic probability matches the target distribution. After multiple iterations, the particle glare viewing angle logic data is generated. In each step of the logic learning process, the error is calculated by the cross entropy function, and the gradient step size is dynamically adjusted based on the learning rate to ensure the accuracy of data fitting.

[0081] Step S4: Design an automated recognition firmware based on the particle glare viewing angle logic data to obtain the particle glare viewing angle recognition firmware, and send the particle glare viewing angle recognition firmware to the cloud platform to perform construction engineering environmental monitoring.

[0082] In the embodiment of the present invention, based on the logic data of the particle glare angle, an automatic identification firmware is designed. The design process of the identification firmware includes three stages: logic circuit optimization, algorithm integration and embedded system development. First, according to the weight matrix extracted from the logic data, a hardware description language (such as Verilog) is used to design an FPGA-based logic circuit. The circuit is used to process the multispectral data stream in real time and identify the changing trend of the glare angle. Next, the designed logic circuit is integrated with the glare angle detection algorithm to optimize the operation efficiency through hardware acceleration. In the development of the embedded system, the firmware is connected to the acquisition module of the multispectral sensor using a serial communication interface, and the operation of each module is coordinated using an independent clock signal. The firmware test verifies its functionality and accuracy by simulating the multispectral data stream generated by the scene. Finally, the designed firmware is uploaded to the cloud platform through the hardware interface. On the cloud platform, the firmware receives the multispectral data sent back from the construction site in real time and performs glare monitoring and analysis to ensure the efficient operation of the construction engineering environment monitoring system.

[0083] Preferably, step S1 comprises the following steps:

[0084] Step S11: deploying a multispectral imaging sensor at the construction site, and collecting real-time spectral data of dust particles to obtain a real-time spectrum of the particles;

[0085] Step S12: performing baseline correction processing on the real-time spectrum of the particulate matter to obtain a real-time baseline-corrected spectrum of the particulate matter;

[0086] Step S13: removing noise from the particle real-time baseline correction spectrum to obtain the particle real-time baseline correction denoised spectrum;

[0087] Step S14: extracting features from the real-time baseline-corrected and denoised spectrum of the particles to obtain a real-time feature spectrum of the particles.

[0088] In an embodiment of the present invention, at a construction site, multispectral imaging sensors are arranged at multiple key positions based on the characteristics of the construction area and the dynamic characteristics of the distribution of dust particles. The installation height and angle of each sensor are calibrated by a laser rangefinder to ensure that the optical axis of the sensor is perpendicular to the main dust flow direction of the construction area. The sensor separates the spectral data through a multi-band filter group, and decomposes the optical signal from the visible light to the near-infrared region into multiple band signals. The data acquisition process uses a high-sensitivity photodetector to convert the received optical signal into a digital signal, and transmits it to the data acquisition terminal in real time through a high-speed data interface. Through high-frequency sampling technology, at least 1,000 sets of spectral data are collected per second to form a real-time spectrum of particles. The data is stored in a two-dimensional matrix form, with each row representing a time point and each column representing the light intensity value of different wavelengths. The collected real-time spectrum of particles is first subjected to baseline correction processing. The baseline correction process adopts a polynomial fitting method. The overall trend of the baseline is determined by performing a piecewise linear fitting on the low-frequency region of the original spectral data. Then, the baseline is smoothed using a cubic spline interpolation method to eliminate errors caused by discontinuous changes. On this basis, the corrected baseline signal is subtracted from the original spectral signal to obtain the corrected spectral data. This process can effectively eliminate the interference of background signals and ambient light and highlight the characteristic peaks of the particle spectrum. The corrected data is saved again in a two-dimensional matrix form for subsequent processing. The real-time baseline correction spectrum of particles is de-noised to improve the signal-to-noise ratio of the data. First, the correction spectrum is decomposed by wavelet transform method, and the spectral data is decomposed into wavelet coefficients of different scales. For the high-frequency noise part, a threshold is set for soft threshold processing, and the coefficients less than the threshold are set to zero to remove the noise component. Subsequently, the processed wavelet coefficients are reconstructed into a denoised spectrum using an inverse wavelet transform. Furthermore, in order to eliminate the interference of spike noise, the denoised spectrum is smoothed by a median filter, and the window size is dynamically adjusted according to the resolution of the spectral data. Finally, the real-time baseline correction denoised spectrum of particles is obtained, and the data integrity and smoothness are significantly improved. Using the real-time baseline correction denoised spectrum of particles, the spectral feature information of particles is obtained by feature extraction technology. The feature extraction process includes peak recognition, peak width calculation and peak area measurement. The local extremum search algorithm is used to locate the peak in the spectrum, and the peak accuracy is determined by quadratic polynomial fitting. Next, the peak shape is modeled using the Gaussian curve fitting method to determine the peak width and peak area. In order to improve the stability of feature extraction, the entire spectrum is normalized to eliminate feature offsets caused by differences in sampling intensity. Feature data is stored in vector form, and each vector contains information such as peak wavelength, peak height, peak width, and peak area, forming a real-time feature spectrum of particles, providing a data basis for subsequent particle type analysis.

[0089] Preferably, step S2 comprises the following steps:

[0090] Step S21: Acquire real-time environmental humidity data and environmental light intensity change data;

[0091] Step S22: simulating the spectrum color conversion effect of the real-time characteristic spectrum of the particulate matter according to the real-time humidity data of the environment to obtain the spectrum color conversion effect data;

[0092] Step S23: performing scattered glare viewing angle analysis on the spectral color conversion effect data based on the ambient light intensity change data to obtain scattered glare viewing angle data;

[0093] Step S24: determining the medium influence weight of the real-time characteristic spectrum of the particle according to the scattered glare viewing angle data, and obtaining the particle glare influence weight data.

[0094] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0095] Step S21: Acquire real-time environmental humidity data and environmental light intensity change data;

[0096] In an embodiment of the present invention, at a construction site, an environmental sensor array is deployed to obtain real-time data on changes in ambient humidity and light intensity. Humidity data is collected through a capacitive humidity sensor. The capacitance value of the sensor changes with humidity. The changing capacitance signal is converted into a digital signal through a modulation and demodulation circuit and recorded as real-time humidity data of the environment. The sampling frequency is set to once per second to capture dynamic changes. Light intensity change data is collected through a photodiode combined with a high-precision analog-to-digital converter. The photodiode generates a current signal in response to changes in light intensity, which is then processed by a low-noise operational amplifier and digitized to generate ambient light intensity change data. Humidity data and light intensity data are stored in the form of time series, respectively. Each record contains a sampling timestamp, humidity percentage, and light intensity value to ensure data traceability and time consistency.

[0097] Step S22: simulating the spectrum color conversion effect of the real-time characteristic spectrum of the particulate matter according to the real-time humidity data of the environment to obtain the spectrum color conversion effect data;

[0098] In an embodiment of the present invention, the real-time humidity data of the environment is used to simulate the spectral color conversion effect of the real-time characteristic spectrum of the particulate matter. First, the humidity data is input into the physical spectrum conversion formula, and the modulation effect of humidity on the spectral absorption characteristics is calculated by the humidity influence model based on the Lambert-Beer law. Humidity changes will cause the thickness of the water film on the surface of the particulate matter to change, thereby changing the absorption and scattering characteristics of the spectrum. The humidity time series is aligned with the sampling time of the spectral data by interpolation to achieve synchronous calculation. Then, matrix operations are used to perform humidity correction on the spectral signal of each wavelength to generate spectral color conversion effect data. The data is stored in the form of a spectral matrix, and each wavelength corresponds to a humidity-corrected spectral intensity value for subsequent analysis.

[0099] Step S23: performing scattered glare viewing angle analysis on the spectral color conversion effect data based on the ambient light intensity change data to obtain scattered glare viewing angle data;

[0100] In an embodiment of the present invention, the spectral color conversion effect data is analyzed for scattered glare viewing angle using the ambient light intensity change data. First, a modulation model of light intensity change on the spectral scattering angle is established based on the Mie scattering theory, and the model is used to calculate the scattering intensity distribution of the spectrum at different viewing angles. The periodic and non-periodic components of the spectral data under light intensity changes are analyzed by discrete Fourier transform, and the change in spectral intensity is decomposed into basic frequency components and harmonic components. Subsequently, the geometric optics method is used to simulate the scattering path of particles at different viewing angles, and the spectral viewing angle characteristics under specific lighting conditions are calculated in combination with the light intensity time series data. Finally, the analysis results are used to generate scattered glare viewing angle data, which is stored in the form of a viewing angle-spectral intensity matrix, where each row represents the intensity distribution of a spectral wavelength and each column represents a different viewing angle.

[0101] Step S24: determining the medium influence weight of the real-time characteristic spectrum of the particle according to the scattered glare viewing angle data, and obtaining the particle glare influence weight data.

[0102] In an embodiment of the present invention, the medium influence weight is determined for the real-time characteristic spectrum of the particulate matter based on the scattered glare viewing angle data. First, the contribution weight of the scattered spectral intensity at different viewing angles is calculated by the weighted average method, and the weight factor is obtained by fitting the relationship curve between the scattered viewing angle and the light intensity. Subsequently, the viewing angle-spectral intensity matrix is ​​subjected to singular value decomposition using matrix decomposition technology to extract the main contribution components of the scattered spectrum to quantify the relative impact of different viewing angles on the spectrum. Combined with the medium optical property model, the refractive index and absorption coefficient of different optical media are introduced into the calculation, and the changes in the spectral characteristics of the particulate matter under the composite medium are analyzed, and finally the particulate matter glare influence weight data is generated. The data is stored in the form of a weight vector, and each weight value corresponds to a spectral wavelength, providing a basis for subsequent spectral data optimization.

[0103] Preferably, step S22 includes the following steps:

[0104] Step S221: calculating the particle size mean difference of the particle real-time characteristic spectrum to obtain the particle size mean difference;

[0105] Step S222: performing particle concentration change trend analysis on the particle real-time characteristic spectrum to obtain particle concentration change trend data;

[0106] Step S223: analyzing the particle humidity adsorption effect on the particle concentration change trend data and the particle size average difference based on the real-time humidity data of the environment to obtain the particle humidity adsorption effect data;

[0107] Step S224: performing light scattering band intensity analysis on the real-time characteristic spectrum of the particles according to the humidity adsorption effect data of the particles to obtain humidity adsorption scattering band intensity data;

[0108] Step S225: simulating the spectrum color conversion effect on the humidity adsorption scattering band intensity data to obtain spectrum color conversion effect data.

[0109] In the embodiment of the present invention, the mean difference of particle size of the real-time characteristic spectrum of the particles is calculated by statistically analyzing the light scattering intensity distribution of the characteristic spectrum. First, according to the Mie scattering theory, there is a functional relationship between the light scattering intensity and the particle size. The light scattering intensity of the characteristic spectrum at multiple wavelengths is grouped and calculated, and the scattering intensity is mapped to the corresponding particle size range according to the preset particle size scattering function model. Subsequently, based on the particle size distribution function, the particle size mean difference index is obtained by calculating the standard deviation and mean ratio of each group of particle sizes. Throughout the process, the numerical integration method is used to integrate the distribution curve of the spectral scattering intensity to ensure the accuracy of the particle size distribution analysis. Finally, the particle size mean difference is stored in vector form, and each data point represents the mean difference value at a specific time for subsequent humidity effect analysis. The particle concentration change trend analysis is performed on the real-time characteristic spectrum of the particles, and the dynamic characteristics of the particle concentration are revealed by the change of the spectral integral intensity of the continuous time series. First, the total scattering intensity of the characteristic spectrum in a specific time window is calculated by the wavelength-by-wavelength integration method as the time series data of the particle concentration. Next, the linear regression analysis method was used to fit the time series and extract the slope and intercept parameters of the concentration change trend. The slope represents the concentration change rate, and the intercept is used as the estimated value of the initial concentration. At the same time, in order to eliminate the noise interference in the spectral signal, the wavelet transform was applied to perform multi-scale decomposition of the concentration time series, retaining only the main trend component and eliminating the influence of high-frequency noise. The analysis results are stored in the form of particle concentration change trend data, including three main fields: time, concentration and change rate. Based on the real-time humidity data of the environment, the particle humidity adsorption effect of the particle concentration change trend data and the particle size mean difference were analyzed. First, the humidity time series data was aligned with the particle concentration change trend data, and the sampling time points of the two sets of data were matched by the interpolation algorithm. Then, based on the Langmuir model of adsorption theory, the influence coefficient of humidity on the adsorption effect of the particle surface was calculated, and the adsorption effect coefficient was obtained by the coupling relationship formula of the particle size mean difference and humidity. Using matrix operations, the difference in adsorption intensity of humidity on particles of different particle sizes was assigned to the particle concentration change trend data to generate the particle humidity adsorption effect data. The data is stored in a three-dimensional matrix, and the three dimensions of the matrix are time, particle size range, and adsorption intensity value. According to the humidity adsorption effect data of the particles, the light scattering band intensity of the real-time characteristic spectrum of the particles is analyzed. First, a specific light scattering band range is selected, and the correction value of the scattering spectrum intensity of each band is calculated in combination with the adsorption intensity value in the humidity adsorption effect data. The integral optical method is used to map the correction of humidity on the optical properties of the particle surface to the spectral intensity distribution curve, thereby generating the scattering band intensity after humidity adsorption. Subsequently, the Fourier transform technology is used to perform spectral decomposition of the band intensity data, extract the main frequency components and harmonic components of the scattering intensity, and analyze the scattering band characteristics under the influence of humidity.Finally, the humidity adsorption scattering band intensity data is stored in a matrix form, where the matrix rows represent the wavelength range and the columns represent the scattering intensity. The humidity adsorption scattering band intensity data is used to simulate the spectral color conversion effect. First, based on the spectral color conversion theory, the functional relationship between the light scattering band intensity and the absorption coefficient and scattering coefficient of a specific wavelength is used to calculate the modulation effect of humidity adsorption on the color characteristics of each wavelength. The piecewise linear interpolation method is used to extend the wavelength of the scattering band intensity data to generate a spectral color distribution curve. On this basis, the spectral data is converted into the CIE color space representation through the color space conversion formula, and the color tristimulus values ​​(XYZ values) are extracted. Finally, the spectral color conversion effect data is generated by combining the correction effects of humidity and particle size on color distribution. The data is stored in a multidimensional array, and each dimension corresponds to the wavelength, humidity and color intensity value.

[0110] Preferably, performing light scattering band intensity analysis on the real-time characteristic spectrum of the particles according to the humidity adsorption effect data of the particles comprises the following steps:

[0111] The particle adsorption capacity is evaluated on the particle humidity adsorption effect data to obtain the particle adsorption capacity data;

[0112] According to the particle adsorption capacity data, the spectrum width change analysis of the real-time characteristic spectrum of the particles is performed to obtain the spectrum width change data of the particles;

[0113] Performing spectral shift measurement on the particle spectral width variation data to obtain spectral shift measurement data;

[0114] The light scattering band intensity is analyzed based on the particle spectral width change data and spectral shift measurement data to obtain the humidity adsorption scattering band intensity data.

[0115] In an embodiment of the present invention, the particle adsorption capacity evaluation is completed by analyzing the humidity dependence characteristics of the particle surface recorded in the humidity adsorption effect data. First, the humidity adsorption effect data is grouped by particle size range, and the cumulative value of the adsorption intensity of each group of particle size particles is calculated by numerical integration method to obtain the total adsorption amount of particles in different particle size ranges. Subsequently, the adsorption capacity per unit volume of the particles under the current humidity conditions is calculated in combination with the adsorption isotherm theory (such as Langmuir or BET adsorption model). The calculation is achieved by fitting the relationship curve between the adsorption isotherm and the humidity adsorption effect data, and the fitting parameters are optimized using the least squares method. The evaluation results are stored in the form of particle adsorption capacity data, and the data include the corresponding relationship between the particle size, humidity level and the corresponding adsorption capacity value. The spectrum width change analysis is completed by evaluating the modulation of the adsorption capacity on the spectral absorption peak and scattering peak morphology. First, according to the particle adsorption capacity data, the real-time characteristic spectrum is divided into multiple sub-intervals according to the wavelength range, and the peak position, full width at half maximum (full width at halfmaxima) and peak intensity of each interval are calculated. Subsequently, combined with the variation law of the adsorption capacity of particles, the width variation trend of the spectral absorption peak and scattering peak under different humidity conditions was analyzed. The contribution coefficient of adsorption capacity to the change of spectral width was quantified through multivariate regression analysis, and the results were stored in matrix form. Finally, the spectral width variation data of particles recorded the width variation value of each wavelength interval and its corresponding relationship with adsorption capacity for subsequent translation measurement calculation. The spectral translation measurement was completed by analyzing the systematic shift of the spectral peak position with humidity conditions. First, the peak position recorded in the spectral width variation data was extracted as a time series, and the principal component analysis (PCA) method was used to remove the cross-effect of width variation on position shift, and only the principal component information related to humidity adsorption was retained. Subsequently, the spectral translation model caused by adsorption capacity was constructed by calculating the dynamic shift of the spectral peak position and its correlation coefficient with the humidity adsorption capacity. To ensure the accuracy of the translation measurement, the discrete Fourier transform was applied to perform frequency domain analysis on the shift data, extract the main shift components and calculate the corresponding intensity change amplitude. Finally, the spectral translation measurement data recorded the dynamic shift values ​​and their change rates of different wavelengths. The light scattering band intensity analysis is completed by integrating the width change data and the translation measurement data to finely decompose the effect of humidity adsorption on the overall morphology of the spectrum. First, the width change data and the translation measurement data are aligned based on the wavelength, and the sampling points of the two sets of data are jointly interpolated using the Lagrange interpolation method to generate a preliminary distribution curve of the scattering intensity of the entire band. Subsequently, based on the Rayleigh scattering and Mie scattering theories, the modulation effect of humidity adsorption on the scattering intensity of each band is calculated, and the optical density formula is applied to correct the results. Finally, the Gaussian distribution fitting technique is used to fit the scattering intensity distribution of each band in sections to extract the main scattering peak positions and intensity parameters.The humidity adsorption scattering band intensity data is stored as a multidimensional array, which contains wavelength, intensity, and humidity adsorption related parameters to support further analysis.

[0116] Preferably, step S23 includes the following steps:

[0117] Step S231: calculating the ultraviolet intensity period variance of the ambient light intensity change data to obtain ultraviolet intensity period variance data;

[0118] Step S232: performing ultraviolet wavelength switching frequency analysis on the ambient light intensity variation data according to the ultraviolet intensity period variance data to obtain ultraviolet wavelength switching frequency data;

[0119] Step S233: obtaining the real-time location of construction workers on the construction project;

[0120] Step S234: performing scattered glare enhancement calculation on the spectrum color conversion effect data according to the ultraviolet wavelength switching frequency data to obtain spectrum color scattered glare enhancement data;

[0121] Step S235: performing scattered glare viewing angle analysis on the real-time position of construction workers on the construction project based on the spectral color scattered glare enhancement data to obtain scattered glare viewing angle data.

[0122] In an embodiment of the present invention, the calculation of the ultraviolet intensity period variance is completed by statistically analyzing the time series of the ultraviolet band in the ambient light intensity change data. First, the light intensity change data is segmented according to the time dimension, and the intensity value of the ultraviolet wavelength (280nm to 400nm) in each time period is extracted. The extracted time series is decomposed in the frequency domain by fast Fourier transform (FFT) to determine the periodic component of the ultraviolet intensity. Then, the variance of the intensity value in each cycle is calculated, and the mean and standard deviation of the variance of each cycle are statistically calculated to form a period variance index of the ultraviolet intensity. During the data processing process, the sliding window method is used to smooth the time series to reduce the influence of random noise. The ultraviolet intensity period variance data is stored in the form of an array, including the variance value of each cycle and its distribution characteristics. The ultraviolet wavelength switching frequency analysis is achieved by dynamically tracking the intensity changes of different ultraviolet wavelengths in the ambient light intensity change data. According to the period variance data generated in step S231, the peak and valley values ​​of the intensity fluctuation within the main cycle range are locked, and the switching point of the wavelength during the peak and valley change process is identified. The slope of the light intensity change curve is used as the evaluation index of the wavelength switching frequency through the polynomial fitting method. The adaptive filtering algorithm is used to eliminate the interference of high-frequency noise on the wavelength switching frequency to ensure the accuracy of the results. Finally, the ultraviolet wavelength switching frequency data is stored as a multi-dimensional matrix, including wavelength, time and switching frequency values. The real-time position acquisition of construction workers is achieved through a multi-base station positioning system installed in key areas of the construction site. The system uses ultra-wideband (UWB) positioning technology to perform two-way time of flight (TOF) measurements between the positioning tags worn by construction workers and the base station to calculate the three-dimensional coordinate position of the construction workers in real time. The data of each positioning tag is sent to the central processing unit through high-frequency wireless communication, and the positioning data is converted into coordinates in combination with the building plan. To ensure accuracy, the Kalman filtering algorithm is used to dynamically optimize the positioning data to remove the influence of multipath effects and signal interference. The real-time position data of construction workers is stored in a table form, including timestamps, three-dimensional coordinates and tag numbers. The calculation of scattered glare enhancement is achieved by integrating ultraviolet wavelength switching frequency data with spectral color conversion effect data. First, the wavelength switching frequency data is mapped to the wavelength range corresponding to the spectral conversion effect data, and the ultraviolet wavelength switching frequency is normalized using a weighting factor to balance the scattering effects of different wavelengths. Then, the scattering enhancement coefficient of each band is calculated by combining the Rayleigh scattering and Mie scattering theories and using the relationship between ultraviolet intensity and wavelength switching frequency. The enhancement coefficient is multiplied by the original spectral color conversion effect data through matrix operations to generate spectral data after enhanced scattered glare. Finally, the spectral color scattered glare enhancement data is recorded in the form of a multidimensional tensor, which contains the enhanced scattering intensity distribution. The scattered glare viewing angle analysis is completed by spatially geometrically mapping the spectral color scattered glare enhancement data with the real-time orientation data of the construction personnel.First, based on the three-dimensional model of the construction site, the spectral scattering intensity distribution is aligned with the positioning data to establish a spatial relationship model between the scattered light path and the perspective of the construction workers. Ray tracing technology is used to calculate the viewing angle range and intensity of each construction worker affected by the scattered glare in the current position. In order to accurately describe the glare distribution, spherical harmonic functions are used to model the scattered light field to obtain the brightness gradient within the perspective of the construction worker. Finally, the scattered glare perspective data is recorded in a matrix form, including the viewing angle range, affected bands and scattering intensity values ​​of each construction worker, which is used to guide construction safety management.

[0123] Preferably, the scattered glare enhancement calculation is performed on the spectrum color conversion effect data according to the ultraviolet wavelength switching frequency data by using a scattered glare intensity calculation formula, wherein the scattered glare intensity calculation formula is as follows:

[0124] ;

[0125] In the formula, Indicates the scattered glare intensity value, represents the mean value of the ultraviolet wavelength switching frequency, represents a natural constant, Indicates the degree of dispersion of ultraviolet wavelengths. Represents the color saturation of the spectral color conversion effect data, Indicates the wavelength range value of the ultraviolet wavelength switching frequency data, Indicates the central wavelength of the UV wavelength switching frequency data, The scattering attenuation index representing the switching frequency data of the ultraviolet wavelength, Indicates the UV dispersion coefficient in the UV wavelength switching frequency data, Represents the error correction value of the scattered glare intensity calculation formula.

[0126] The present invention provides a formula for calculating the intensity of scattered glare, which comprehensively considers multiple variables, making the calculation of scattered glare more comprehensive and able to reflect the impact of different factors on spectral color conversion. , represents the overall intensity of ultraviolet wavelengths in the spectrum. By using the mean, the influence of different wavelengths can be better balanced; the natural constant , used in the formula for exponential functions, enhances the nonlinear characteristics of the spectral response and can more accurately simulate the behavior of light in real environments; the degree of ultraviolet wavelength dispersion , reflects the change of ultraviolet wavelength, affects the scattering characteristics of light, and can accurately model the response of light of different wavelengths; color saturation of spectral color conversion effect data , which affects the effect of spectral color conversion. The higher the saturation, the stronger the color perception of light, which directly affects the perception of scattered glare; the wavelength range value of ultraviolet wavelength switching frequency data , describes the distribution width of the ultraviolet wavelength switching frequency data. The larger the range, the more complex the spectral response, and can effectively simulate a variety of ambient light conditions; the central wavelength of the ultraviolet wavelength switching frequency data , which represents the main characteristic wavelength of the spectrum, has a direct impact on the scattering effect and can help determine the most sensitive wavelength region; the scattering attenuation index of the ultraviolet wavelength switching frequency data , which indicates the intensity attenuation of light due to scattering during propagation, affecting the final intensity of scattered glare; the ultraviolet dispersion coefficient in the ultraviolet wavelength switching frequency data , describes the dispersion characteristics in the ultraviolet wavelength switching frequency data, affects the width and distribution of the spectrum, and helps to simulate the characteristics of various light sources; error correction value of the scattered glare intensity calculation formula , used to correct potential errors in the formula and ensure the accuracy and reliability of the calculation results.

[0127] Preferably, performing scattered glare viewing angle analysis on the real-time position of construction workers on a construction project based on spectral color scattered glare enhancement data comprises the following steps:

[0128] Analyze the visual movement range based on the real-time position of construction workers in the construction project to obtain the visual movement range of the workers;

[0129] Calculate the scattering angle of the spectral color scattering glare enhancement data to obtain the spectral color scattering angle data;

[0130] According to the spectral color scattering angle data and the spectral color scattering glare enhancement data, the vision movement glare stimulation range of the personnel is matched to obtain the vision movement glare stimulation range;

[0131] The scattered glare visual angle analysis is performed on the visual motion glare stimulation range to obtain the scattered glare visual angle data.

[0132] In the embodiment of the present invention, the real-time position data of the construction personnel are synchronously acquired by a laser rangefinder and a high-precision GPS positioning device. Combined with the site layout, the real-time positioning point is mapped to the two-dimensional plane coordinate system of the construction site using a coordinate transformation method. On this basis, a line of sight projection algorithm is used to construct the range of the visible area according to the facial orientation of the personnel and the preset visual level range. The definition of the visual movement range is divided into multiple subdivided sectors by a discretized angle step, and the maximum boundary position in these areas is calculated respectively. The data is finally stored in the format of a polygonal boundary to describe the area of ​​the visual coverage range. The key parameters in the spectral color scattering glare enhancement data include the wavelength of the light source, the position of the scattering center and the intensity distribution. The Henry-Greenstein phase function widely used in optical scattering theory is used to define the scattering angle range and the directivity factor. By analyzing the spatial geometric relationship between the light source position and the construction site, the projection angle range of the glare in the construction area is calculated. The scattering angle calculation adopts a ray tracing method to gradually iterate the propagation path of the light beam in a multi-point environment, and finally obtain the effective scattering angle distribution data corresponding to each light source. The spectral color scattering angle data is combined with the directionality of the light source intensity in the spectral color scattering glare enhancement data to match the stimulation range. The matching algorithm is based on the principle of geometric area intersection, and the polygonal boundary of the construction worker's visual movement range is superimposed and calculated with the effective area of ​​the scattered light. During the superposition process, the Monte Carlo method is used to sample and verify the area intersection to ensure the accuracy and stability of the calculation. The calculation result is expressed as the effective area stimulated by glare within the construction worker's visual coverage range, which is stored in a multi-dimensional matrix data format. The visual movement glare stimulation range data is combined with the on-site geometric layout data, and the distribution of glare in the visual field is accurately located by spatial vector decomposition technology. The weighted angle integration method is used to calculate the contribution of glare intensity to the viewing angle in each direction. In particular, segmented integration is performed within the upper and lower limits of the glare angle to identify the main impact angle of glare on the construction worker. The analysis results are output in the form of angle values ​​of the viewing angle range, and the glare intensity level in the key direction is marked. The final scattered glare viewing angle data is stored as a vector set of multi-angle distribution, which is convenient for subsequent environmental monitoring and adjustment.

[0133] Preferably, step S3 comprises the following steps:

[0134] Step S31: normalizing the particle glare impact weight data to obtain particle glare impact weight normalized data;

[0135] Step S32: performing convolution calculation on the particle glare influence weight normalized data to obtain particle glare influence weight convolution data;

[0136] Step S33: Based on the policy gradient algorithm, logic learning is performed on the particle glare impact weight convolution data and the scattered glare viewing angle data to obtain the particle glare viewing angle logic data.

[0137] In the embodiment of the present invention, the particle glare impact weight data is quantitative data reflecting the degree of influence of different particles on light scattering and glare. These data are normalized to map the data to a unified scale for subsequent calculation normalization. The normalization process uses a standardized formula. The core idea of ​​the step is to ensure that the processed data is between 0 and 1 by subtracting the minimum value and dividing by the range (maximum value minus minimum value), thereby eliminating the influence of the scale difference of different particle data. This processing step is to ensure that there will be no calculation errors due to inconsistent data scales during subsequent data processing, and to maintain the balance and comparability of the data. The convolution operation is a basic operation in signal processing, which aims to combine the glare impact weight of particles with the convolution kernel to obtain a new weight distribution. The specific method of convolution kernel selection is based on the scattering characteristics of particles and ambient lighting conditions. The purpose of convolution is to smooth the data through a filter and extract the influence pattern of particles on scattered glare. This process combines the particle impact at each moment with the weight of the surrounding moments by weighted superposition to calculate the overall contribution of particles to glare. In the convolution calculation, a convolution kernel with Gaussian distribution characteristics is selected to effectively filter out random noise and retain the trend change of the particle effect. Through this convolution operation, the effect of particles on scattered glare is further optimized, and a smooth weight distribution is obtained, which is convenient for subsequent further analysis and processing. The policy gradient algorithm is a reinforcement learning algorithm used to learn the optimal behavior strategy in an unknown environment. In this step, the policy gradient algorithm learns the optimal logical model to predict the effect of particles on scattered glare by jointly analyzing the particle glare effect weight convolution data and the scattered glare perspective data. First, the particle glare effect weight convolution data and the scattered glare perspective data are used as input states. The goal of the algorithm is to learn the optimal "strategy" taken in each state, that is, how to adjust the particle effect weight to reduce the impact of glare on construction workers. The policy gradient algorithm continuously optimizes the weights and updates the strategy through the gradient ascent method, so that the results obtained by the strategy can maximize the reduction of the impact of glare, and finally form a set of logical rules, that is, a reasonable matching scheme for the effect of particles on scattered glare. The core idea of ​​the policy gradient algorithm is to continuously optimize the objective function and adjust the strategy through multiple interactions with the environment. In the specific implementation, the policy function is trained using the stochastic gradient descent method (SGD), and the weight of each step is continuously adjusted until it converges to an optimal strategy to obtain the particle glare perspective logic data. This data reflects the influence of different particles on scattered glare and can be used for subsequent building environment monitoring optimization.

[0138] Preferably, the present invention further provides a construction engineering environment monitoring system for executing the construction engineering environment monitoring method as described above, the construction engineering environment monitoring system comprising:

[0139] The spectral feature extraction module is used to deploy multi-spectral imaging sensors at the construction site, collect real-time spectral data of dust particles, and obtain real-time spectra of particles; extract features from the real-time spectra of particles to obtain real-time feature spectra of particles;

[0140] The scattered glare analysis module is used to simulate the spectral color conversion effect of the real-time characteristic spectrum of the particulate matter to obtain the spectral color conversion effect data; perform scattered glare viewing angle analysis on the spectral color conversion effect data to obtain scattered glare viewing angle data; determine the medium influence weight based on the scattered glare viewing angle data to obtain the particulate matter glare influence weight data;

[0141] A logic learning module is used to perform convolution calculation on the particle glare impact weight data to obtain particle glare impact weight convolution data; based on the strategy gradient algorithm, logic learning is performed on the particle glare impact weight convolution data and the scattered glare viewing angle data to obtain particle glare viewing angle logic data;

[0142] The firmware design module is used to design automatic recognition firmware based on the particle glare perspective logic data, obtain the particle glare perspective recognition firmware, and send the particle glare perspective recognition firmware to the cloud platform to perform construction project environmental monitoring.

[0143] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0144] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A construction engineering environment monitoring method, characterized in that: The following steps are involved: Step S1: deploying a multispectral imaging sensor at a construction site, and collecting real-time spectral data of dust particles to obtain a real-time spectrum of the particles; extracting features from the real-time spectrum of the particles to obtain a real-time feature spectrum of the particles; Step S2: simulating the spectral color conversion effect of the real-time characteristic spectrum of the particulate matter to obtain spectral color conversion effect data; performing scattered glare viewing angle analysis on the spectral color conversion effect data to obtain scattered glare viewing angle data; determining the medium influence weight according to the scattered glare viewing angle data to obtain particulate matter glare influence weight data; Step S3: performing convolution calculation on the particle glare impact weight data to obtain particle glare impact weight convolution data; performing logic learning on the particle glare impact weight convolution data and the scattered glare viewing angle data based on the strategy gradient algorithm to obtain particle glare viewing angle logic data; Step S4: Design an automated recognition firmware based on the particle glare viewing angle logic data to obtain the particle glare viewing angle recognition firmware, and send the particle glare viewing angle recognition firmware to the cloud platform to perform construction engineering environmental monitoring.

2. The construction engineering environment monitoring method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: deploying a multispectral imaging sensor at the construction site, and collecting real-time spectral data of dust particles to obtain a real-time spectrum of the particles; Step S12: performing baseline correction processing on the real-time spectrum of the particulate matter to obtain a real-time baseline-corrected spectrum of the particulate matter; Step S13: removing noise from the particle real-time baseline correction spectrum to obtain the particle real-time baseline correction denoised spectrum; Step S14: extracting features from the real-time baseline-corrected and denoised spectrum of the particles to obtain a real-time feature spectrum of the particles.

3. The construction engineering environment monitoring method according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: Acquire real-time environmental humidity data and environmental light intensity change data; Step S22: simulating the spectrum color conversion effect of the real-time characteristic spectrum of the particulate matter according to the real-time humidity data of the environment to obtain the spectrum color conversion effect data; Step S23: performing scattered glare viewing angle analysis on the spectral color conversion effect data based on the ambient light intensity change data to obtain scattered glare viewing angle data; Step S24: determining the medium influence weight of the real-time characteristic spectrum of the particle according to the scattered glare viewing angle data, and obtaining the particle glare influence weight data.

4. The construction engineering environment monitoring method according to claim 3, characterized in that: Step S22 includes the following steps: Step S221: calculating the particle size mean difference of the particle real-time characteristic spectrum to obtain the particle size mean difference; Step S222: performing particle concentration change trend analysis on the particle real-time characteristic spectrum to obtain particle concentration change trend data; Step S223: analyzing the particle humidity adsorption effect on the particle concentration change trend data and the particle size average difference based on the real-time humidity data of the environment to obtain the particle humidity adsorption effect data; Step S224: performing light scattering band intensity analysis on the real-time characteristic spectrum of the particles according to the humidity adsorption effect data of the particles to obtain humidity adsorption scattering band intensity data; Step S225: simulating the spectrum color conversion effect on the humidity adsorption scattering band intensity data to obtain spectrum color conversion effect data.

5. The construction engineering environment monitoring method according to claim 4, characterized in that: The light scattering band intensity analysis of the real-time characteristic spectrum of particles based on the humidity adsorption effect data of particles includes the following steps: The particle adsorption capacity is evaluated on the particle humidity adsorption effect data to obtain the particle adsorption capacity data; According to the particle adsorption capacity data, the spectrum width change analysis of the real-time characteristic spectrum of the particles is performed to obtain the spectrum width change data of the particles; Performing spectral shift measurement on the particle spectral width variation data to obtain spectral shift measurement data; The light scattering band intensity is analyzed based on the particle spectral width change data and spectral shift measurement data to obtain the humidity adsorption scattering band intensity data.

6. The construction engineering environment monitoring method according to claim 3, characterized in that: Step S23 includes the following steps: Step S231: calculating the ultraviolet intensity period variance of the ambient light intensity change data to obtain ultraviolet intensity period variance data; Step S232: performing ultraviolet wavelength switching frequency analysis on the ambient light intensity variation data according to the ultraviolet intensity period variance data to obtain ultraviolet wavelength switching frequency data; Step S233: obtaining the real-time location of construction workers on the construction project; Step S234: performing scattered glare enhancement calculation on the spectrum color conversion effect data according to the ultraviolet wavelength switching frequency data to obtain spectrum color scattered glare enhancement data; Step S235: Perform scattered glare viewing angle analysis on the real-time position of construction workers on the construction project based on the spectral color scattered glare enhancement data to obtain scattered glare viewing angle data.

7. The construction engineering environment monitoring method according to claim 6, characterized in that: The scattered glare enhancement calculation is performed on the spectrum color conversion effect data according to the ultraviolet wavelength switching frequency data, and is performed through the scattered glare intensity calculation formula, wherein the scattered glare intensity calculation formula is as follows: ； In the formula, Indicates the scattered glare intensity value, represents the mean value of the ultraviolet wavelength switching frequency, represents a natural constant, Indicates the degree of dispersion of ultraviolet wavelengths. Represents the color saturation of the spectral color conversion effect data, Indicates the wavelength range value of the ultraviolet wavelength switching frequency data, Indicates the central wavelength of the UV wavelength switching frequency data, The scattering attenuation index representing the switching frequency data of the ultraviolet wavelength, Indicates the UV dispersion coefficient in the UV wavelength switching frequency data, Represents the error correction value of the scattered glare intensity calculation formula.

8. The construction engineering environment monitoring method according to claim 6, characterized in that: The scattered glare perspective analysis of the real-time position of construction workers in construction projects based on spectral color scattered glare enhanced data includes the following steps: Analyze the visual movement range based on the real-time position of construction workers in the construction project to obtain the visual movement range of the workers; Calculate the scattering angle of the spectral color scattering glare enhancement data to obtain the spectral color scattering angle data; According to the spectral color scattering angle data and the spectral color scattering glare enhancement data, the vision movement glare stimulation range of the personnel is matched to obtain the vision movement glare stimulation range; The scattered glare visual angle analysis is performed on the visual motion glare stimulation range to obtain the scattered glare visual angle data.

9. The construction engineering environment monitoring method according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: normalizing the particle glare impact weight data to obtain particle glare impact weight normalized data; Step S32: performing convolution calculation on the particle glare influence weight normalized data to obtain particle glare influence weight convolution data; Step S33: Based on the policy gradient algorithm, logic learning is performed on the particle glare impact weight convolution data and the scattered glare viewing angle data to obtain the particle glare viewing angle logic data.

10. A construction engineering environment monitoring system, characterized in that: Used to execute the construction engineering environment monitoring method as claimed in claim 1, the construction engineering environment monitoring system comprises: The spectral feature extraction module is used to deploy multi-spectral imaging sensors at the construction site, collect real-time spectral data of dust particles, and obtain real-time spectra of particles; extract features from the real-time spectra of particles to obtain real-time feature spectra of particles; The scattered glare analysis module is used to simulate the spectral color conversion effect of the real-time characteristic spectrum of the particulate matter to obtain the spectral color conversion effect data; perform scattered glare viewing angle analysis on the spectral color conversion effect data to obtain scattered glare viewing angle data; determine the medium influence weight based on the scattered glare viewing angle data to obtain the particulate matter glare influence weight data; A logic learning module is used to perform convolution calculation on the particle glare impact weight data to obtain particle glare impact weight convolution data; based on the strategy gradient algorithm, logic learning is performed on the particle glare impact weight convolution data and the scattered glare viewing angle data to obtain particle glare viewing angle logic data; The firmware design module is used to design automatic recognition firmware based on the particle glare perspective logic data, obtain the particle glare perspective recognition firmware, and send the particle glare perspective recognition firmware to the cloud platform to perform construction project environmental monitoring.

Images