Multi-terminal collaborative road illumination comprehensive evaluation method and system

Abstract

The invention discloses a multi-terminal collaborative road illumination comprehensive evaluation method and system. The method comprises the steps of constructing a street lamp geographic coordinate database; acquiring street lamp illuminance values, image gray features and timestamps collected by detection equipment on different vehicles; identifying and rejecting the interference spectrum to obtain detection data of multiple vehicles without interference; on the basis of a correlation coefficient between the street lamp illuminance value and the image average gray value and a spectrum relative coefficient corresponding to the street lamp illuminance value, verifying the detection data of the multiple vehicles subjected to interference removal so as to determine effective detection data of the multiple vehicles; determining the weight of the effective detection data of the multiple vehicles based on the street lamp geographic coordinate database; combining the effective detection data of the multiple vehicles with the corresponding weights to determine a comprehensive illumination value; and determining the state of the street lamp according to comparative analysis of the comprehensive illumination value and historical data so as to update a street lamp state file. By implementing the method provided by the invention, accurate monitoring and early warning of performance degradation of the street lamp can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to a road illumination comprehensive evaluation method, more particularly to a multi-terminal cooperative road illumination comprehensive evaluation method and system. BACKGROUND

[0002] In today's increasingly popular smart city concept, efficient lighting detection technology is also a necessary component of building an intelligent city management system. However, the current lighting detection technology has several limitations, mainly reflected in the limitations of single sensor detection, the isolation of single detection, and the disconnection between actual visual perception and numerical measurement. First, the traditional illumination sensor detection method is easily disturbed by environmental light, such as oncoming vehicle lights or building lighting, which increases the possibility of misjudgment. Second, the main drawback of existing technology is the isolation of single detection, including defects in the time dimension and deficiencies in the spatial dimension. Specifically, single detection can only provide information on the instantaneous state and cannot capture intermittent faults that may exist in street lights, such as start-up delay or power fluctuations. In addition, due to the lack of statistical analysis of historical data for the same section, it is difficult to establish a reliable evaluation benchmark for street light status, making the evaluation results less accurate.

[0003] Furthermore, existing solutions fail to fully utilize the data correlation and cumulative analysis capabilities of multi-vehicle cooperation, which means they miss the opportunity to improve detection accuracy through large sample statistics. At the same time, current detection methods mostly focus on the measurement of illumination values, while ignoring the important influence of light source distribution uniformity on driver visual perception. Due to the lack of multi-vehicle perspective fusion mechanism, these methods cannot comprehensively evaluate the actual lighting effect of street lights, especially at different angles and different times.

[0004] Finally, regarding the evaluation of street light status, existing technology lacks continuity and gradualness representation. The decay and failure of street lights usually occur gradually, but existing technology is difficult to accurately track this change process. In particular, there is a lack of an effective multi-vehicle long-term tracking mechanism to construct a time series model of street light performance degradation, and there is also a lack of sufficient correlation analysis means when dealing with detection data at different time points for the same section. The existence of these problems limits the improvement of street light maintenance efficiency and service quality, indicating the need for a new comprehensive solution to overcome the above challenges.

[0005] Therefore, it is necessary to design a new method based on multi-vehicle cooperative multi-sensor and visual fusion technology to realize accurate monitoring and early warning of street light performance degradation, in order to solve the problem of existing technology lacking multi-vehicle cooperative monitoring mechanism and comprehensive spatio-temporal data analysis, resulting in large accidental errors in street light status detection, inability to effectively capture the gradual degradation process of brightness, and single angle of visual evaluation. SUMMARY

[0006] The present application aims to overcome the defects of the prior art, and provide a multi-terminal cooperative road illumination comprehensive evaluation method and system.

[0007] To achieve the above-mentioned purpose, the present application adopts the following technical solutions: a multi-terminal cooperative road illumination comprehensive evaluation method, comprising: constructing a street lamp geographic coordinate database; obtaining street lamp illumination values, image gray scale features and time stamps collected by detection devices on different vehicles to obtain multi-vehicle detection data; identifying and eliminating interference spectra in the multi-vehicle detection data to obtain multi-vehicle detection data with interference removed; verifying the multi-vehicle detection data with interference removed based on a correlation coefficient between street lamp illumination values and image average gray scale values and a spectrum relative coefficient corresponding to the street lamp illumination values to determine valid multi-vehicle detection data; determining weights of the valid multi-vehicle detection data based on the street lamp geographic coordinate database; determining a comprehensive illumination value based on the valid multi-vehicle detection data and the corresponding weights; determining a state of a street lamp based on a comparative analysis of the comprehensive illumination value and historical data to update a street lamp state archive.

[0008] Further technical solutions thereof are as follows: the identification and elimination of the interference spectra in the multi-vehicle detection data to obtain the multi-vehicle detection data with interference removed, comprising: calculating a spectrum similarity in the multi-vehicle detection data, confirming real signals for detection data corresponding to a similarity meeting a requirement, confirming interference signals for detection data corresponding to a similarity not meeting the requirement, eliminating the interference signals to obtain the multi-vehicle detection data with interference removed.

[0009] Further technical solutions thereof are as follows: the verification of the multi-vehicle detection data with interference removed based on a correlation coefficient between street lamp illumination values and image average gray scale values and a spectrum relative coefficient corresponding to the street lamp illumination values to determine valid multi-vehicle detection data, comprising: verifying effectiveness of the multi-vehicle detection data with interference removed based on a correlation coefficient between street lamp illumination values and image average gray scale values to determine first valid data; calculating a spectrum relative coefficient corresponding to the street lamp illumination values for the first valid data to determine valid multi-vehicle detection data.

[0010] Further technical solutions thereof are as follows: the verification of effectiveness of the multi-vehicle detection data with interference removed based on a correlation coefficient between street lamp illumination values and image average gray scale values to determine first valid data, comprising: For each vehicle, a correlation coefficient between the street light illuminance value and the image average gray value is calculated based on the detection data using the ratio of covariance to standard deviation to measure the internal consistency of the single vehicle detection data; for multiple vehicles, the correlation coefficient between the street light illuminance value and the image average gray value is calculated based on the detection data within the same time period, and the correlation coefficients calculated by each vehicle are compared to measure the consistency and reliability of the detection data of multiple vehicles, and the first effective data is determined; Wherein, when the correlation coefficient corresponding to the vehicle deviates from the correlation coefficients corresponding to other vehicles, the weight corresponding to the vehicle is reduced.

[0011] A further technical solution is that the first effective data is used to calculate a frequency spectrum relative coefficient corresponding to the street light illuminance value to determine the effective detection data of multiple vehicles, including: The product of the frequency spectrum response and the exponentially decaying interference intensity function within the current time window is calculated for the first effective data to obtain the frequency spectrum relative coefficient corresponding to the street light illuminance value; The effective detection data of multiple vehicles is determined using the relationship between the frequency spectrum relative coefficient corresponding to the street light illuminance value and the set threshold.

[0012] A further technical solution is that the weight of the effective detection data of the multiple vehicles is determined based on the street light geographic coordinate database, including: When different vehicles pass through the same place, the detection data of the multiple vehicles is automatically recognized and associated by a GPS matching algorithm based on the street light geographic coordinate database; In a specific time window, the weight is assigned according to the time sequence of the effective detection data of the multiple vehicles to obtain a time weight; The weight is assigned according to the distance between the vehicle and the standard detection point of the street light to obtain a spatial weight; The weight is assigned according to the historical detection accuracy of the vehicle to obtain a credibility weight.

[0013] A further technical solution is that the street light illuminance value, image gray feature, and time stamp collected by the detection equipment on different vehicles are obtained to obtain the detection data of multiple vehicles, including: The sliding window length is automatically adjusted according to the traffic density, and the sampling frequency of the key road section is adjusted.

[0014] A further technical solution is that the effective detection data of the multiple vehicles is combined with the corresponding weight to determine a comprehensive illuminance value, including: The effective detection data of each vehicle is weighted and summed with the weight to obtain a weighted detection value; The weighted detection values of all vehicles are averaged to obtain a street lamp instantaneous state value; The street lamp instantaneous state value, the historical trend analysis value and the multiple detection consistency evaluation value are combined according to respective weight coefficients to obtain a comprehensive illumination value.

[0015] Further technical solutions of the application are as follows: the state of the street lamp is determined according to comparison of the comprehensive illumination value and historical data, and the state of the street lamp is updated to update a street lamp state archive, including: A judgment threshold is determined according to the historical data and the current environmental factors; the judgment threshold is dynamically adjusted by combining the average value and the standard deviation statistically calculated from the historical data and the weighted sum of the environmental influence factors, and is continuously updated by using a sliding window mechanism to adapt to environmental changes to obtain the judgment threshold; The state of the street lamp is determined according to comparison of the comprehensive illumination value and the judgment threshold, and the state of the street lamp is updated to update a street lamp state archive.

[0016] The application further provides a multi-terminal cooperative road illumination comprehensive evaluation system, including: A construction unit is configured to construct a street lamp geographic coordinate database; An acquisition unit is configured to acquire street lamp illumination values, image gray scale features and time stamps collected by detection devices on different vehicles to obtain detection data of multiple vehicles; An identification and elimination unit is configured to identify and eliminate interference spectra in the detection data of the multiple vehicles to obtain detection data of the multiple vehicles after interference is removed; An effectiveness determination unit is configured to verify the detection data of the multiple vehicles after interference is removed based on a correlation coefficient between the street lamp illumination values and the image average gray scale values and a relative coefficient of the spectrum corresponding to the street lamp illumination values to determine effective detection data of the multiple vehicles; A weight determination unit is configured to determine weights of the effective detection data of the multiple vehicles based on the street lamp geographic coordinate database; A fusion unit is configured to combine the effective detection data of the multiple vehicles to determine a comprehensive illumination value; A comparison and analysis unit is configured to determine the state of the street lamp according to comparison of the comprehensive illumination value and historical data to update a street lamp state archive.

[0017] Compared with the prior art, the present application has the beneficial effects that: the present application constructs a street lamp geographic coordinate database, combines the collected illumination values, image gray scale features and time stamp information from detection equipment of different vehicles, uses multi-vehicle cooperative multi-sensor and visual fusion technology, identifies and eliminates interference spectrum in the detection data, verifies the validity of the data based on the correlation coefficient between the illumination value and the average image gray scale value and the relative coefficient of the spectrum corresponding to the street lamp illumination value, and calculates the comprehensive illumination value according to the corresponding weight given according to the geographic position of the street lamp. Through comparison and analysis with historical data, the state of the street lamp is accurately judged and the state archive is updated, so that the precise monitoring and early warning of the performance degradation of the street lamp are realized. The method solves the problems of lack of multi-vehicle cooperative monitoring mechanism and insufficient spatio-temporal data analysis in the prior art, effectively reduces accidental errors, captures the gradual change process of brightness attenuation, overcomes the limitation of single visual evaluation angle, and provides a more comprehensive and accurate street lamp state detection scheme.

[0018] The present application will be further described below in combination with the drawings and specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0020] Figure 1 The flowchart of the multi-terminal cooperative road illumination comprehensive evaluation method provided by the embodiment of the present application is shown in the figure. Figure 2 The schematic diagram of the detection data of the multi-vehicle provided by the embodiment of the present application is shown in the figure. Figure 1 ; Figure 3 The schematic diagram of the detection data of the multi-vehicle provided by the embodiment of the present application is shown in the figure. Figure 2 ; Figure 4 The schematic diagram of the detection data of the multi-vehicle provided by the embodiment of the present application is shown in the figure. Figure 3 ; Figure 5 The schematic diagram of the fusion process provided by the embodiment of the present application is shown in the figure. Figure 6 The schematic diagram of the comprehensive illumination value provided by the embodiment of the present application is shown in the figure. Figure 7 The schematic diagram of the historical data provided by the embodiment of the present application is shown in the figure. Figure 8 The schematic diagram of the brightness change when the vehicle starts detection provided by the embodiment of the present application is shown in the figure. Figure 9A schematic block diagram of a multi-terminal cooperative road illumination comprehensive evaluation system provided by an embodiment of the present application is shown in FIG. 1. Figure 10 A schematic block diagram of a computer device provided by an embodiment of the present application is shown in FIG. 2. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0022] It should be understood that, when used in the specification and the appended claims, the terms "comprise" and "include" indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0023] It should also be understood that the terms used in the present application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms, unless the context clearly indicates otherwise.

[0024] It should be further understood that the term "and / or" used in the present application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0025] Please refer to Figure 1 , Figure 1A schematic flowchart of a multi-terminal cooperative road illumination comprehensive evaluation method provided by an embodiment of the present application. The multi-terminal cooperative road illumination comprehensive evaluation method is applied to a server. The server interacts with detection devices of different vehicles and terminals, collects data such as street lamp illumination values, image gray scale features and time stamps by integrating detection devices on multiple vehicles, and uses frequency spectrum analysis to eliminate interference signals, thereby ensuring the accuracy and reliability of the data. The method is based on a street lamp geographic coordinate database, uses a GPS matching algorithm to associate data of different vehicles at the same location, and combines time weight, space weight and credibility weight to perform weighted processing on effective detection data to determine a comprehensive illumination value. By comparing the comprehensive illumination value with historical data, the judgment threshold is dynamically adjusted and the street lamp status archive is updated, thereby achieving accurate monitoring and early warning of street lamp performance degradation. This method solves the problems of large accidental error, inaccurate brightness attenuation capture and single visual evaluation angle in the prior art due to the lack of a multi-vehicle cooperative monitoring mechanism and comprehensive space-time data analysis, thereby greatly improving the accuracy and reliability of street lamp status detection.

[0026] Figure 1 A flowchart of a multi-terminal cooperative road illumination comprehensive evaluation method provided by an embodiment of the present application. As shown in Figure 1 the method includes the following steps S110 to S170.

[0027] S110, constructing a street lamp geographic coordinate database.

[0028] In this embodiment, the street lamp geographic coordinate database refers to a systematic data collection that records the precise geographic position information and related attributes of each street lamp in the detection area. Each street lamp is assigned a unique ID and corresponds to an accurate GPS coordinate, which enables accurate identification and positioning of each street lamp. Constructing such a database is the basis for implementing the multi-terminal cooperative road illumination comprehensive evaluation method, and provides the necessary prerequisites for subsequent data collection, space-time matching and multi-source data fusion.

[0029] Specifically, constructing the street lamp geographic coordinate database includes the following steps: First, a detailed on-site investigation of all street lamps in the target area is needed to record the position information of each street lamp. This information can be directly measured by high-precision GPS equipment or obtained from existing street lamp position data from city planning or municipal departments.

[0030] The collected street lamp position information is entered into the database management system, and each street lamp is assigned a unique identifier (ID). This ID helps to distinguish different street lamps and is an important basis for subsequent data association and analysis.

[0031] In addition to basic geographic location information, a series of attributes can be defined for each streetlight, such as installation date, bulb type, power, etc. These additional information helps to have a more comprehensive understanding of the working status of streetlights and their performance trend.

[0032] The streetlight geographic coordinate database needs to be regularly maintained and updated to reflect any newly added, removed or renovated streetlights. This usually involves cooperation with municipal management departments to ensure that the information in the database is always up-to-date.

[0033] Finally, the built streetlight geographic coordinate database will be integrated with other systems such as detection equipment on vehicles, central servers, etc., to support multi-vehicle collaborative road illumination evaluation. Through GPS matching algorithm, detection data from different vehicles at the same location can be automatically identified and associated, thus establishing a multi-source detection file for the streetlight, providing solid data support for subsequent comprehensive analysis.

[0034] In summary, the construction of streetlight geographic coordinate database is crucial for efficient and accurate road lighting state monitoring, which not only improves the efficiency of data processing, but also provides strong technical support for precise positioning and solving streetlight faults.

[0035] S120, acquiring streetlight illumination values, image gray scale features and time stamps collected from detection equipment on different vehicles to obtain multi-vehicle detection data.

[0036] In this embodiment, multi-vehicle detection data refers to the information related to streetlight lighting conditions collected from different vehicles equipped with specific detection equipment, including illumination sensors, cameras, GPS positioning modules, etc. These data are obtained through multi-vehicle collaboration, aiming to fuse and analyze data at different time points and spatial locations to more accurately evaluate the working status of streetlights. Specifically, multi-vehicle detection data mainly includes the following contents: Streetlight illumination value: This is the data directly measured by the illumination sensor installed on the vehicle, indicating the light intensity of the area where the streetlight is located at a certain time. The value is usually in lux units, and for the same streetlight, the measured illumination values may differ due to different positions or angles of different vehicles.

[0037] Image gray scale feature: The camera captures the scene in front of the windshield, especially the part containing the streetlight, and then extracts the gray scale feature of the image through image processing technology. This step is mainly to reflect the brightness of the road surface under the driver's visual perception. The gray scale feature can be quantified by calculating the average gray value of the pixels in the image, which provides another perspective to evaluate the actual lighting effect of the streetlight in addition to the illumination value.

[0038] Timestamp: Each data point is accompanied by an accurate timestamp, identifying the specific time of data collection. This is crucial for subsequent time series analysis, as it allows the system to arrange data from different vehicles in chronological order for comparison and fusion analysis.

[0039] Geographical Position Information: The exact location of the vehicle at each data collection is recorded using a GPS positioning module. This is essential for determining which car detected which streetlight at which location. Geographical position information not only helps to correlate data from different vehicles but also serves as the basis for constructing a database of streetlight geographical coordinates.

[0040] Data Fusion and Verification: After receiving data from multiple vehicles, the system first performs preliminary processing, including removing obvious outliers and disturbances. Then, based on the principles of time weight distribution, spatial weight distribution, and credibility weight distribution mentioned above, the data is weighted and averaged to generate a comprehensive state evaluation value. This process also includes the calculation of the correlation between illumination values and image grayscale values to further verify the effectiveness of the data.

[0041] In summary, in this embodiment, "multi-vehicle detection data" is not just a series of isolated numerical sets, but a complex, multi-level information system that includes physical measurement results (such as illumination values), environmental perception information (such as image grayscale features), time dimensions (such as timestamps), and spatial positions (such as geographical position information). Through scientific and reasonable data fusion and analysis methods, these data can be effectively converted into accurate assessments of streetlight working conditions, supporting the intelligent upgrade of urban lighting management systems.

[0042] In this embodiment, the sliding window length is automatically adjusted according to traffic density, and the sampling frequency of the region is adjusted in combination with key road segments.

[0043] Specifically, the change in traffic density during different time periods affects the quality of streetlight state detection data. During peak hours, vehicles are dense, and data updates quickly; during off-peak hours, vehicles are sparse, and data updates slowly. Therefore, a mechanism needs to be designed to dynamically adjust the sliding window length to adapt to different traffic density situations.

[0044] Real-time traffic information for each road segment is obtained through a GPS positioning module and a wireless communication module.

[0045] Based on historical data analysis, the average traffic density during different time periods is analyzed, and a threshold (such as high density, normal, and low density) is set.

[0046] When traffic density is high, a shorter sliding window (e.g., 2 minutes) is used to capture finer-grained time variation characteristics, ensuring rapid response; In normal traffic density, use a medium-length sliding window (e.g., 5 minutes) to balance timeliness and stability; If the traffic density is low, extend the sliding window to a longer period (e.g., 10 minutes or more) to ensure sufficient data for analysis.

[0047] Some urban areas may have higher requirements for street lighting, such as business districts, schools, or accident-prone areas. These areas are referred to as "critical sections" and require higher monitoring accuracy and faster response speed.

[0048] Determine which areas belong to critical sections based on city planning documents, traffic accident reports, and resident feedback.

[0049] For known critical sections, the system will prioritize scheduling more vehicles for data collection or increase the data upload frequency of vehicles passing through the area.

[0050] Additional fixed sensor equipment is also installed to supplement the data from mobile devices.

[0051] Dynamic adjustment mechanism: In addition to pre-set critical sections, the system should have learning ability to dynamically adjust the critical section list and corresponding sampling frequency according to the latest road conditions (such as new construction, temporary activities, etc.).

[0052] Through the above measures, this system not only maintains efficient data processing capability under different traffic conditions, but also provides enhanced service support for special needs. The specific performance is: Flexible adjustment of parameter settings according to actual conditions reduces data bias caused by a single standard.

[0053] Especially in critical sections, problems can be found and solved faster, improving public safety.

[0054] Avoids excessive consumption of computing resources in unnecessary cases, achieving effective allocation of resources.

[0055] In summary, this adaptive adjustment scheme based on traffic density and critical section characteristics greatly improves the flexibility and practicality of the entire street lighting monitoring system, providing strong technical support for smart city management.

[0056] The limitation of traditional single vehicle detection is that its sliding window is limited to the driving time and speed of a single vehicle, resulting in limited data volume and possible lack of comprehensiveness. In contrast, multi-vehicle coordination provides an innovative breakthrough by establishing a dynamic sliding window mechanism based on multi-vehicle time series, enabling the acquisition of more comprehensive time dimension data.

[0057] When vehicle A detects a certain streetlight within 0-5 minutes, and vehicle B also detects the same streetlight within 3-8 minutes, the system will fuse the data of both vehicles within the overlapping time period of 3-5 minutes for verification and analysis.

[0058] The length of the sliding window is dynamically adjusted according to the number density of vehicles passing through. For example, a shorter window (e.g., 2 minutes) is used when traffic is heavy to achieve higher temporal resolution, while a longer window (e.g., 10 minutes) is used when traffic is sparse to ensure sufficient data volume is collected.

[0059] The system can identify which road segments are particularly important for monitoring and prompt the operator to increase the sampling frequency of vehicles in these areas, thereby improving the data collection density and quality in these areas.

[0060] In this way, by utilizing the data of multiple vehicles at different time periods, the system not only tracks the dynamic trend between lighting intensity and gray scale, but also achieves continuous monitoring through the relay between vehicles, which cannot be achieved by a single vehicle.

[0061] S130, identifying and removing interference spectrum in the detection data of the plurality of vehicles to obtain the detection data of the plurality of vehicles after removing interference.

[0062] In this embodiment, the detection data of the plurality of vehicles after removing interference refers to identifying and removing interference signals caused by non-streetlight light sources (such as oncoming vehicle lights, LED advertising screens, etc.) through a series of spectrum analysis and consistency verification steps, so as to retain data that truly reflects the working state of the streetlight.

[0063] Calculate the spectral similarity in the detection data of the plurality of vehicles, confirm the detection data corresponding to the similarity meeting the requirements as a real signal, confirm the detection data corresponding to the similarity not meeting the requirements as an interference signal, remove the interference signal, and obtain the detection data of the plurality of vehicles after removing interference.

[0064] Specifically, this process includes the following key steps: First, extract the frequency domain components of the illumination signal from the data collected by each participating vehicle. This step uses Fourier transform or other spectrum analysis methods to convert time domain signals to frequency domain representation.

[0065] Next, the system compares the spectral data collected by different vehicles for the same streetlight within the same time period. By calculating the similarity between these spectra (e.g., using Pearson correlation coefficient or cosine similarity), the matching degree between them can be quantified.

[0066] A threshold of spectral similarity is determined to distinguish which data is the real streetlight signal and which is the abnormal signal caused by the interference source. If the similarity of a group or multiple groups of data is higher than the threshold, it is considered that these data reflect the real situation; otherwise, it is considered as interference.

[0067] Once the data set below the similarity threshold is identified, it is marked as an interference signal and removed from the overall data set. This process helps to reduce the risk of misjudgment due to local interference and improve the accuracy of the final evaluation results.

[0068] In order to further ensure the reliability of the data, a "voting mechanism" can also be used in some cases. This means that for each potential main frequency component, only when the majority of vehicles (more than a certain proportion) report similar results, it will be officially recognized as part of the valid signal. This method can effectively exclude accidental interference received by individual vehicles.

[0069] In summary, the detection data of multiple vehicles after removing interference involves complex spectral analysis techniques and the application of statistical principles. Real streetlight signals should be consistent among multiple vehicles; interference signals usually have spatial limitations; if multiple vehicles detect abnormal spectrum at the same time, it may be a real streetlight flicker problem; through multi-vehicle statistics, local interference received by individual vehicles can be identified and excluded. Through this method, valuable information can be accurately separated from massive raw data, providing a solid foundation for efficient urban lighting management.

[0070] S140, verifying the detection data of multiple vehicles after removing interference based on the correlation coefficient between the streetlight illuminance value and the image average gray value and the spectral relative coefficient corresponding to the streetlight illuminance value, to determine the valid detection data of multiple vehicles.

[0071] In this embodiment, the valid detection data of multiple vehicles refers to the data set that can accurately reflect the relationship between the streetlight illuminance and the image average gray value of the environment where each vehicle is located after verification and screening, ensuring the consistency and reliability of the data, which can be used for subsequent analysis and processing.

[0072] In an embodiment, the step S140 described above can include steps S141-S142.

[0073] S141, verifying the validity of the detection data of multiple vehicles after removing interference based on the correlation coefficient between the streetlight illuminance value and the image average gray value, to determine the first valid data.

[0074] In this embodiment, the first valid data refers to a data set determined by measuring the internal consistency of single vehicle and multi-vehicle detection data based on the correlation coefficient between the street light illumination value and the image average gray value, excluding data that may be disturbed or abnormal, ensuring the quality of the data.

[0075] For each vehicle, the correlation coefficient between the street light illumination value and the image average gray value is calculated based on the detection data using the ratio of covariance to standard deviation to measure the internal consistency of single vehicle detection data; for multiple vehicles, the correlation coefficient between the street light illumination value and the image average gray value is calculated based on the detection data using the ratio of covariance to standard deviation within the same time period, and the correlation coefficients calculated by each vehicle are compared to measure the consistency and reliability of the detection data of multiple vehicles, to determine the first valid data. Wherein, when the correlation coefficient corresponding to the vehicle deviates from the correlation coefficients corresponding to other vehicles, the weight corresponding to the vehicle is reduced.

[0076] Specifically, for each vehicle participating in detection, the street light illumination value (Lux) and the image average gray value collected by the vehicle are analyzed. The linear relationship strength between the two is calculated by the ratio of covariance to standard deviation (i.e. Pearson correlation coefficient formula). This step helps to measure the consistency of the internal data of a single vehicle. Within the same time period, the above calculation process is repeated for the data of all vehicles to obtain the correlation coefficient of each vehicle. The correlation coefficients calculated by different vehicles are compared. If the correlation coefficient of a vehicle deviates significantly from other vehicles (e.g. deviation exceeds 0.3), it is considered that the vehicle may be affected by local interference or equipment failure, thus reducing its weight.

[0077] In this embodiment, by simultaneously detecting the same light source (such as street light) from multiple angles by different vehicles, the matching degree of the illumination sensor data and the visual perception can be cross-verified. This cross-vehicle data collection and analysis method effectively eliminates accidental errors and local interference in single vehicle detection, improving the accuracy and reliability of the detection results.

[0078] For each vehicle participating in detection, first calculate the correlation coefficient between its illumination value and the image average gray value. This process involves using the ratio of covariance to standard deviation to quantify the linear relationship strength between the two. This step is the foundation, which helps us understand how the change in light intensity at a certain location affects the trend of image brightness change without external interference.

[0079] After obtaining the individual correlation coefficients for each vehicle, the next step is to perform multi-vehicle consistency verification. This step involves comparing the correlation coefficients of different vehicles within the same time period. If the correlation coefficients of all vehicles remain at a high level (e.g., all greater than 0.8), it can be considered that the street lighting effect in that area is good, and there is good consistency between the illuminance sensor data and visual perception.

[0080] The system also incorporates an anomaly detection mechanism to handle potential data bias. When the correlation coefficient of a vehicle deviates significantly from that of other vehicles (e.g., a deviation exceeding 0.3), the system considers that vehicle's data to be interfered with, and automatically reduces the weight of that data. This effectively reduces the impact of outlier data on the overall analysis results, further enhancing the system's robustness and environmental adaptability.

[0081] In the absence of external interference, i.e., under normal circumstances, as the intensity of street lighting increases, the average gray value of the image should also increase accordingly, showing a stable linear positive correlation.

[0082] However, in the presence of interfering sources (such as oncoming headlights or LED advertising screens), although the illuminance sensor may detect higher brightness values, the overall grayscale of the image will not increase accordingly. This inconsistency can be quantified using a correlation coefficient, helping to identify and filter out these outlier data points.

[0083] In summary, implementing a multi-vehicle cross-validation mechanism not only eliminates accidental interference factors but also improves the overall system's detection accuracy, enabling it to better adapt to different environmental conditions. This method provides reliable data support and technical assurance for fields such as intelligent transportation systems and urban lighting management.

[0084] S142. Calculate the spectral relative coefficient corresponding to the street light illuminance value for the first valid data to determine the valid detection data for multiple vehicles.

[0085] In one embodiment, step S142 described above may include steps S1421 to S1422.

[0086] S1421. For the first valid data, calculate the product of the spectral response and the exponentially decaying interference intensity function within the current time window to obtain the spectral relative coefficient corresponding to the street light illuminance value.

[0087] In this embodiment, the spectral relative coefficient corresponding to the street light illuminance value is an index calculated by performing spectral analysis on the first valid data and combining it with an exponentially decaying interference intensity function. It is used to evaluate the characteristics and anti-interference ability of these data in the frequency domain and further confirm the validity of the data.

[0088] Specifically, first, a Fast Fourier Transform (FFT) is performed on each group of illumination value sequences in the first valid data to obtain its spectral response.

[0089] A function that decays exponentially over time is defined where represents the interference intensity, and k is the decay rate constant. In this embodiment, the interference intensity is essentially a weighting factor, which works as follows: No or slight interference: when the interference intensity is close to 0, it indicates that the data has been hardly affected by external factors. At this time, the value of the weighting factor is close to 1, meaning that the original spectral response of this data will be preserved to the greatest extent. This is because, in the absence of interference, the data has higher reliability and accuracy.

[0090] Severe interference: on the contrary, when the interference intensity significantly increases, it indicates that the data may have suffered from severe external interference. In this case, the value of the weighting factor will tend to 0, effectively reducing the influence of this unreliable data in subsequent analysis. In this way, those data points with greater interference can be suppressed or penalized, avoiding their negative impact on the overall analysis results.

[0091] To quantify the interference intensity, a method based on the consistency of inter-vehicle data is adopted. The specific steps are as follows: First, the system will analyze the data recorded by each vehicle (such as illumination values) and calculate the correlation coefficient between them. This process aims to assess the consistency and reliability of the data.

[0092] Next, the system will compare the data of a single vehicle with the average level of other vehicles in the same area. If the data (whether it is the illumination value or the correlation coefficient) of a vehicle deviates significantly from the average of other vehicles, the degree of deviation can be regarded as an indicator of the local interference intensity of the data of that vehicle.

[0093] Finally, according to the above deviation degree, the system can determine the specific interference intensity of each vehicle's data and adjust its corresponding weighting factor accordingly. This allows the quality of the entire data set and the credibility of the analysis results to be improved through dynamic adjustment of the weights even in complex environments.

[0094] In summary, by introducing a weighting factor that adjusts with the intensity of interference, we can more accurately reflect the true value of the data in the frequency domain analysis, reduce the error caused by interference, and ultimately improve the overall performance and reliability of data analysis. This method not only considers the characteristics of the data itself, but also takes into account the impact of environmental factors on data quality, providing a more scientific and reasonable data processing strategy.

[0095] By multiplying the spectrum response of each vehicle with the corresponding interference intensity function, we obtain a correlation coefficient that reflects the frequency domain characteristics. This coefficient not only considers the frequency components of the signal itself, but also considers the degree of influence of potential interference. Here, represents the spectrum response within a specific time window.

[0096] S1422, determine the effective detection data of multiple vehicles using the relationship between the spectrum relative coefficient corresponding to the street light illumination value and the set threshold.

[0097] According to historical data analysis and experimental results, set a reasonable spectrum relative coefficient threshold value corresponding to the street light illumination value .

[0098] Compare the spectrum relative coefficient corresponding to the street light illumination value of each vehicle with its corresponding threshold value. If , it is considered that the detection data of this vehicle is valid; otherwise, it is marked as interference data.

[0099] Combine the first effective data obtained in the previous step with the results of this time based on the spectrum relative coefficient corresponding to the street light illumination value, and comprehensively determine the effective detection data set of multiple vehicles. This set is used for subsequent street light status evaluation, trend analysis, and warning mechanism triggering applications.

[0100] Through these two key steps, the system can effectively filter out high-quality effective information from complex multi-source data, ensuring the accuracy and reliability of the street light monitoring system.

[0101] After successfully removing all interference signals, only the remaining high-quality data will be used for subsequent analysis. This includes but is not limited to time weight allocation, spatial weight allocation, and credibility weight allocation operations to generate an accurate description of the current working status of a specific street light. Finally, the system updates the corresponding street light status archive based on the processed data, recording the latest maintenance requirements or fault information. This is of great significance for city management departments to take timely action to repair damaged street lights.

[0102] In this embodiment, the correlation coefficient between the street light illumination value and the image average gray value is to compare the relationship between the light intensity measured by the illumination sensor and the average gray value of the image captured by the camera. The purpose is to verify whether there is a stable linear positive correlation between the two, so as to confirm whether the street lighting effect is good.

[0103] This method can effectively exclude accidental errors and local interference in single vehicle detection, because different vehicles detect the same street light from multiple angles at the same time, providing more comprehensive data verification.

[0104] The spectral relative coefficient corresponding to the street light illumination value mainly focuses on the frequency characteristic difference of different light sources (such as street light source, vehicle light, LED advertising screen, etc.). By analyzing the spectral components of these light sources, non-street light source interference can be identified and suppressed, and further consistency analysis can be performed to determine the effectiveness of the data.

[0105] S150, determining the weight of the valid detection data of the multiple vehicles based on the street light geographic coordinate database.

[0106] In this embodiment, the weight refers to the different importance degrees given to the detection data points according to factors such as the time freshness of the data, the accuracy of the spatial position, and the credibility of the data provider, so as to optimize the result accuracy after multi-source data fusion.

[0107] In an embodiment, the above step S150 can include steps S151-S154.

[0108] S151, when different vehicles pass through the same place, based on the street light geographic coordinate database, the detection data of the multiple vehicles is automatically identified and associated by GPS matching algorithm.

[0109] In this embodiment, when different vehicles pass through the same street light position, the system uses GPS matching algorithm to identify these vehicles and associate their detection data with the corresponding street light ID, forming a multi-source detection file of the street light.

[0110] This step ensures that even data from different vehicles can be correctly classified to a specific street light, laying the foundation for subsequent data fusion.

[0111] S152, according to the time sequence of the valid detection data of the multiple vehicles, the weight is assigned to obtain the time weight.

[0112] In this embodiment, the time weight refers to giving different weight values according to the freshness of the detection data. Generally, the newer the data is considered more reliable, so the weight is higher.

[0113] In a set time window (e.g. 30 minutes before and after), all detection data of passing vehicles are assigned weights according to their detection time. For example, a linear decreasing function or an exponential decay function can be used to calculate the time weight of each data point.

[0114] S153, assign a weight according to the distance between the vehicle and the standard detection point of the streetlight to obtain a spatial weight.

[0115] In this embodiment, the spatial weight is a weight assigned based on the distance between the vehicle and the standard detection point of the streetlight. In theory, the closer the data point, the higher the reliability.

[0116] For each detection data point, a weight is calculated according to its actual distance from the standard detection point of the streetlight. For example, an inverse function or a Gaussian distribution function can be used to reflect the influence of distance on weight.

[0117] S154, assign a weight according to the historical detection accuracy of the vehicle to obtain a credibility weight.

[0118] In this embodiment, the credibility weight reflects the accuracy level of the historical detection records of each vehicle. Those vehicles that have performed well in the past will be given a higher weight.

[0119] The error rate or accuracy rate in the past detection tasks of each vehicle is analyzed, and a weight reflecting its reliability is assigned for each new detection. This can be achieved through statistical models or machine learning algorithms, such as calculating the mean absolute error (MAE) or mean square error (MSE), and then converting it into a corresponding weight value.

[0120] After completing the above four steps, the system will calculate the time weight, spatial weight and credibility weight for each detection data point. Then, through a weighted fusion algorithm, these weights are combined to form the final comprehensive weight, which is used to generate the comprehensive state evaluation value of the streetlight.

[0121] Through this method, the system not only improves the accuracy of single detection, but also uses the diverse information provided by multiple vehicles to overcome the limitations of a single perspective, thus providing more comprehensive and reliable streetlight state evaluation results.

[0122] S160, determine the comprehensive illumination value by combining the effective detection data of the multiple vehicles with the corresponding weights.

[0123] In the present embodiment, the comprehensive illumination value refers to a single numerical value representing the lighting state of the street lamp at a certain moment, which is obtained by performing weighted fusion processing on the effective detection data collected by multiple vehicles. This process not only considers the time, spatial position, and credibility weight of each vehicle detection data, but also combines historical trend analysis and consistency evaluation of multiple detections to provide an evaluation result that comprehensively reflects the actual working state of the street lamp.

[0124] In an embodiment, the above-mentioned step S160 can include steps S161-S163.

[0125] S161, the effective detection data of each vehicle is weighted and summed with the weight to obtain a weighted detection value.

[0126] In the present embodiment, the weighted detection value refers to an adjusted numerical value reflecting the detection result of each vehicle, which is calculated by assigning appropriate weights to the effective detection data provided by each vehicle according to its time freshness, distance from the street lamp, and historical accuracy rate of the vehicle.

[0127] For the effective detection data (including illumination value, image gray feature, etc.) provided by each vehicle, appropriate weights are assigned according to its time freshness, distance from the street lamp, and historical accuracy rate of the vehicle, and these data are multiplied by their corresponding weights and summed to obtain the weighted detection value for the vehicle. This step ensures that the data of vehicles that are fresher, closer to the standard detection point, and have better historical performance have higher influence.

[0128] S162, the weighted detection values of all vehicles are averaged to obtain a street lamp instantaneous state value.

[0129] In the present embodiment, the street lamp instantaneous state value refers to a single numerical value representing the lighting effect of the street lamp at the current moment, which is obtained by averaging the weighted detection values of all vehicles.

[0130] Next, the system aggregates the weighted detection values of all vehicles and calculates their average value to obtain an instantaneous state value representing the lighting effect of the street lamp at the current moment. In this way, the data of different vehicles is effectively integrated together, reducing the uncertainty caused by errors of individual vehicles.

[0131] S163, the street lamp instantaneous state value, historical trend analysis value, and multiple detection consistency evaluation value are combined according to their respective weight coefficients to obtain a comprehensive illumination value.

[0132] Finally, the system combines the instantaneous state value of the street lamp obtained above with the historical trend analysis value based on long-term monitoring data and the consistency evaluation value between multiple detections, and generates a final comprehensive illumination value by weighting and merging according to the pre-set weight coefficients. This multi-dimensional consideration method can more accurately capture the changes in the working state of the street lamp, discover potential problems in time, and improve the reliability of the overall detection system.

[0133] Through such a process, the system not only provides accurate street lamp lighting state evaluation, but also effectively identifies abnormal situations such as local interference or equipment failure, thereby supporting more scientific and reasonable maintenance decisions.

[0134] In this embodiment, the detection data weighting fusion formula of a single vehicle is as follows: ; is the weighted detection value of the ith vehicle to the nth street lamp; is the time weight (according to the newness or oldness of the detection time); is the spatial weight (according to the distance from the standard monitoring point); is the credibility weight (according to the historical accuracy rate of the vehicle); is the street lamp illumination value of the ith vehicle.

[0135] The multi-vehicle data time sequence fusion is realized by the following formula: This formula calculates the instantaneous state value of the nth street lamp, taking into account the data of all participating vehicles (total number N) and their respective weight factors.

[0136] The comprehensive state evaluation model is used to comprehensively evaluate the state of the street lamp, and its formula is as follows: ; is the current instantaneous state value based on multi-vehicle weighted fusion; is the trend analysis value based on historical time series; is the consistency evaluation value of multiple detection data; , used to adjust the proportion of each item in the final evaluation.

[0137] This state evaluation model can effectively identify the decay of the street lamp. It not only uses current effective data, but also compares historical data to identify long-term changes in illumination trends, and ensures the consistency of multiple detection results on the same section. In this way, the city lighting system can be more accurately maintained and managed.

[0138] S170, determining the state of the street lamp according to the comparison of the comprehensive illumination value and the historical data, to update the street lamp state profile.

[0139] In this embodiment, the street lamp state profile refers to an electronic document recording the historical detection data of each street lamp and its state change (normal, pre-warning or failure). These profiles not only help to monitor the performance and health condition of individual street lamps, but also provide valuable data support for the entire lighting network.

[0140] In an embodiment, the step S170 described above can include steps S171-S172.

[0141] S171, determining a judgment threshold according to the historical data and the current environmental factors; wherein the judgment threshold is dynamically adjusted by combining the average value and the standard deviation of the historical data statistics and the weighted sum of the environmental impact factors, and continuously updated using a sliding window mechanism to adapt to environmental changes.

[0142] In this embodiment, the judgment threshold refers to In this step, the system first collects relevant environmental parameters, including but not limited to weather conditions, traffic density, seasonal changes and time periods, etc. Based on these real-time environmental parameters, the system calculates a comprehensive environmental impact factor , whose formula is as follows: ; is the weather condition factor (sunny = 1.0, cloudy = 1.1, rainy and foggy = 1.2); is the weight value of the weather condition factor; is the traffic density factor (low density = 0.9, normal = 1.0, high density = 1.1); is the weight value of the traffic density factor; is the seasonal factor (spring and summer = 0.95, autumn and winter = 1.05); is the weight value of the seasonal factor; is the time period factor (dusk = 1.1, late night = 1.0, early morning = 0.9); is the weight value of the time period factor.

[0143] The specific values of each factor depend on the actual situation.

[0144] Next, the system dynamically adjusts the historical data using a sliding window mechanism to adapt to environmental changes, and calculates the judgment threshold accordingly. Specifically, the normal threshold and the pre-warning state threshold are calculated according to the following formulas respectively: ; ; where, and are the mean and standard deviation calculated based on historical data within a sliding window, respectively.

[0145] S172, determine the status of the streetlight based on the comparison of the comprehensive illuminance value and the judgment threshold, to update the streetlight status archive.

[0146] At this stage, the system compares the currently calculated comprehensive illuminance value with the previously determined judgment threshold, to determine the status of the streetlight. The specific judgment logic is as follows: Normal state: if ; Warning state: if ; Fault state: if .

[0147] Once the status of the streetlight is determined, the system automatically updates the corresponding streetlight status archive. This step ensures the real-time and accuracy of the streetlight management system, which can timely discover and handle potential problems, thus improving the reliability and efficiency of the public lighting system.

[0148] In addition, in order to further reduce the misjudgment rate, when the system initially judges that there is an abnormality in a streetlight, it will call the data of other vehicles that have recently passed through the section for secondary verification. Only when the results of multiple vehicle detections are consistent and point to an abnormality, will the fault state be finally confirmed. This multi-level verification mechanism effectively improves the accuracy of the detection results and reduces the false positive problem caused by environmental interference.

[0149] In this embodiment, the judgment threshold is actually evaluated by using a threshold adaptive adjustment algorithm to assess the status of the streetlight, thus improving the reliability and efficiency of the public lighting system. The main steps are as follows: The system first collects current environmental parameters, including but not limited to weather conditions, traffic density, seasonal changes, and time periods, etc.

[0150] Based on the collected real-time environmental parameters, a comprehensive environmental impact factor is calculated; The mean and standard deviation of the historical data are updated using a sliding window mechanism to reflect the latest environmental conditions.

[0151] The judgment threshold is dynamically adjusted by combining the weighted sum of the mean and standard deviation of the historical data statistics and the environmental impact factor.

[0152] Compare the comprehensive illuminance value at the current time with the dynamically calculated judgment threshold, and output the status of the streetlight.

[0153] To reduce false positives caused by environmental interference, when the system initially determines that a streetlight has an anomaly, it calls on other vehicles that have recently passed through the section to conduct a secondary verification. Only when multiple vehicle detection results consistently point to an anomaly will the system finally confirm the fault state. This multi-level verification mechanism not only improves the accuracy of the detection results, but also ensures the reliability of the streetlight state evaluation.

[0154] In addition, the system combines trend analysis with the current state to achieve early warning. For repeated detection of the same section, the state score is weighted according to the correlation weight, further improving the accuracy of the detection results. This series of measures effectively ensures the efficient operation and maintenance of the public lighting system.

[0155] The method of this embodiment utilizes different vehicles to detect the same streetlight at different times and locations, obtaining a comprehensive data set. This includes not only temporal differences (e.g., vehicle A passing through a certain streetlight at 8 pm, and vehicle B passing through at 8:15 pm), but also spatial differences (e.g., the main lane is 25 meters away, and the auxiliary lane is 30 meters away).

[0156] Based on the age of the detection data, weights are assigned, with more recent data being considered more accurate and therefore having higher weights. According to the distance between the vehicle and the streetlight, weights are assigned, with closer data points having higher weights to eliminate the limitations of a single observation angle. According to the historical detection performance of the vehicle, weights are assigned, with vehicles that have performed more accurately in the past being given higher weights, thereby improving the reliability of the overall data.

[0157] Through the above weight assignment mechanism, the system can fuse detection data from different vehicles to form a comprehensive state profile for each streetlight.

[0158] To further improve the accuracy of the evaluation, the system not only relies on historical statistical data to set the basic threshold, but also introduces real-time environmental impact factors for adjustment. These environmental factors take into account weather conditions, traffic density, seasonal changes, and time period differences, ensuring that the threshold can be dynamically adjusted according to changes in environmental conditions.

[0159] Compared to traditional single-vehicle static threshold methods, this multi-vehicle collaborative scheme significantly improves the scientificity and accuracy of threshold setting due to the use of more statistical samples and stronger environmental perception capabilities. It can automatically adapt to different environmental conditions, effectively reducing the misjudgment rate and improving the detection reliability of the entire system.

[0160] In summary, this advanced street light condition assessment model, through the integration of multi-vehicle collaborative data collection, precise weight allocation, and dynamic threshold adjustment, provides strong support for the maintenance of public lighting systems, ensuring efficient and reliable operation.

[0161] For example, suppose three cars (A, B, and C) detected the same streetlight at different times and locations, obtaining data such as... Figures 2 to 4 The data shown: Vehicle A: Image grayscale: 102; Illuminance detected: 38 lux; Vehicle B: Image grayscale: 100; Illuminance detected: 36 lux; Vehicle C: Image grayscale: 98; Illuminance detected: 34 lux.

[0162] The system first uses image grayscale values ​​to verify the reliability of the illuminance data. This step is based on the premise that, under normal circumstances, the illuminance value of streetlights is positively correlated with the image grayscale value.

[0163] Calculate the correlation coefficient between illuminance and grayscale in each data set to identify potential interference or outliers. For example, the presence of interfering light sources such as oncoming headlights or LED advertising screens may cause an abnormally high illuminance sensor reading, but the overall grayscale of the image may not change significantly; in this case, the correlation coefficient will decrease significantly.

[0164] In this embodiment, the correlation coefficients of the three sets of data were calculated to be 0.85, 0.87 and 0.83, respectively, all of which are higher than the set threshold of 0.8, indicating that all data were not significantly interfered with and the illuminance values ​​were true and reliable.

[0165] Each set of effective illuminance data is assigned a corresponding weight based on three dimensions: time, space, and credibility.

[0166] Time weighting: More recent data has a higher weight.

[0167] Spatial weight: Data points closer to the target street light have higher weights.

[0168] Credibility weight: Weights are assigned based on the vehicle's historical inspection performance.

[0169] like Figure 5 as well as Figure 6 As shown, using the weighted average algorithm and combining the above weights, the comprehensive illuminance value of the street light is calculated to be 36.2 lux.

[0170] The final evaluation result only outputs the comprehensive illuminance value without including the gray value. This is because the gray value is mainly used for data quality verification, while the road lighting standard (such as CJJ 45-2015) clearly specifies that the evaluation index is in lux.

[0171] In addition, as Figures 7 to 8 indicated, the gray value is easily affected by factors such as camera parameters and exposure settings, and is not suitable as a standardized evaluation index. The illuminance value after multi-vehicle fusion verification has excluded interference and has high reliability.

[0172] The system determines whether the detected data belongs to the same streetlight based on the comparison of the geographic coordinate points of the vehicle when the streetlight is detected.

[0173] The fused comprehensive illuminance value is compared and analyzed with the historical data of the streetlight in the database to update the streetlight status file, realizing the continuous monitoring of the streetlight operating status.

[0174] When the streetlight fails to light up, the problem can be clearly analyzed from the historical data, and abnormal trends can also be found from the continuous data of individual vehicles.

[0175] In this way, multi-vehicle collaborative detection data fusion not only improves the accuracy and reliability of streetlight status evaluation, but also realizes efficient management and maintenance of the public lighting system.

[0176] The method of the embodiment solves the limitations of the prior art in streetlight status evaluation and proposes a new dynamic streetlight status detection method based on multi-vehicle collaboration and multi-sensor-vision fusion. This method integrates multiple technical means to achieve efficient and accurate monitoring and evaluation of streetlight status.

[0177] An illuminance sensor network distributed on multiple mobile vehicles is used to achieve repeated detection of the same streetlight position at multiple time points. This method not only captures changes over different time periods, but also effectively reduces accidental errors caused by single detection.

[0178] Data from different vehicles at the same geographic coordinate point are arranged in chronological order to form a detailed long-term monitoring file. This file helps to understand the historical performance of a specific streetlight and its trends over time.

[0179] Through statistical analysis of the data collected by multiple vehicles, abnormal values or accidental errors can be effectively identified and excluded, thereby improving the reliability and accuracy of the overall detection results.

[0180] A time series model of street light brightness decay is constructed using data from multiple vehicle tracking over a long time span. This enables the system to issue warnings before street light performance begins to degrade, facilitating timely maintenance and avoiding failures.

[0181] Cameras on different vehicles capture images of the same section from multiple angles and extract the gray scale distribution features of these images to establish a more comprehensive and detailed visual assessment model. This method not only provides richer information for state judgment, but also enhances the adaptability to complex environmental factors such as changes in light conditions.

[0182] Through the above measures, a more accurate and forward-looking street light state detection solution is provided, greatly improving the management efficiency and service quality of the public lighting system. In addition, this multi-vehicle cooperative method also provides new ideas and technical support for urban infrastructure monitoring.

[0183] The above multi-terminal cooperative road illumination comprehensive evaluation method, by constructing a street light geographic coordinate database, combining the illumination values, image gray scale features and timestamp information collected from the detection equipment of different vehicles, using multi-vehicle cooperative multi-sensor and visual fusion technology, identifying and removing interference spectrum in the detection data, verifying the validity of the data based on the correlation coefficient between the illumination value and the average gray scale value of the image and the relative coefficient of the spectrum corresponding to the street light illumination value, and according to the geographic location of the street light, giving the corresponding weight, calculating the comprehensive illumination value. By comparing and analyzing with historical data, the state of the street light is accurately judged and the state archive is updated, so as to realize the accurate monitoring and early warning of the performance degradation of the street light. This method solves the problems of lack of multi-vehicle cooperative monitoring mechanism and insufficient spatio-temporal data analysis in the prior art, effectively reduces accidental errors, captures the gradual change process of brightness decay, and overcomes the limitation of single visual evaluation angle, providing a more comprehensive and accurate street light state detection solution.

[0184] Figure 9 is a schematic block diagram of a multi-terminal cooperative road illumination comprehensive evaluation system 300 provided by an embodiment of the present application. As Figure 9 shown, corresponding to the above multi-terminal cooperative road illumination comprehensive evaluation method, the present application also provides a multi-terminal cooperative road illumination comprehensive evaluation system 300. The multi-terminal cooperative road illumination comprehensive evaluation system 300 includes units for executing the above multi-terminal cooperative road illumination comprehensive evaluation method, and the system can be configured in a server. Specifically, please refer to Figure 9 , the multi-terminal cooperative road illumination comprehensive evaluation system 300 includes a construction unit 301, an acquisition unit 302, an identification and removal unit 303, an effectiveness determination unit 304, a weight determination unit 305, a fusion unit 306 and a comparison and analysis unit 307.

[0185] The construction unit 301 is configured to construct a street lamp geographic coordinate database; the acquisition unit 302 is configured to acquire street lamp illumination values, image gray scale features and time stamps collected by detection devices on different vehicles to obtain multi-vehicle detection data; the identification and elimination unit 303 is configured to identify and eliminate interference spectrums in the multi-vehicle detection data to obtain multi-vehicle detection data after interference elimination; the validity determination unit 304 is configured to verify the multi-vehicle detection data after interference elimination based on a correlation coefficient between street lamp illumination values and image average gray scale values and a spectral relative coefficient corresponding to the street lamp illumination values to determine valid multi-vehicle detection data; the weight determination unit 305 is configured to determine weights of the valid multi-vehicle detection data based on the street lamp geographic coordinate database; the fusion unit 306 is configured to determine a comprehensive illumination value by combining the valid multi-vehicle detection data and corresponding weights; and the comparative analysis unit 307 is configured to determine a state of a street lamp based on a comparative analysis of the comprehensive illumination value and historical data to update a street lamp state archive.

[0186] In an embodiment, the identification and elimination unit 303 is configured to calculate spectral similarities in the multi-vehicle detection data, confirm detection data corresponding to similarities meeting requirements as real signals, confirm detection data corresponding to similarities not meeting requirements as interference signals, eliminate the interference signals to obtain multi-vehicle detection data after interference elimination.

[0187] In an embodiment, the validity determination unit 304 includes: A first determination subunit is configured to verify validity of the multi-vehicle detection data after interference elimination based on a correlation coefficient between street lamp illumination values and image average gray scale values to determine first valid data; and a second determination subunit is configured to calculate a spectral relative coefficient corresponding to the street lamp illumination values for the first valid data to determine valid multi-vehicle detection data.

[0188] In an embodiment, the first determination subunit is configured to, for each vehicle, calculate a correlation coefficient between street lamp illumination values and image average gray scale values based on detection data using a ratio of covariance and standard deviation to measure internal consistency of single-vehicle detection data; and for multiple vehicles, calculate a correlation coefficient between street lamp illumination values and image average gray scale values based on detection data using a ratio of covariance and standard deviation in a same time period and compare correlation coefficients calculated by each vehicle to measure consistency and reliability of multi-vehicle detection data to determine first valid data. When a correlation coefficient corresponding to a vehicle deviates from correlation coefficients corresponding to other vehicles, the weight corresponding to the vehicle is reduced.

[0189] In an embodiment, the second determination subunit includes: a coefficient determination module configured to calculate a product of a spectral response in a current time window and an exponential decay interference intensity function for the first effective data to obtain a spectral relative coefficient corresponding to the streetlight illumination value; and a data determination module configured to determine effective detection data of the plurality of vehicles by using a relationship between the spectral relative coefficient corresponding to the streetlight illumination value and a set threshold value.

[0190] In an embodiment, the weight determination unit 305 comprises: an association subunit configured to automatically identify and associate the detection data of the plurality of vehicles by a GPS matching algorithm based on the streetlight geographic coordinate database when different vehicles pass through the same location; a time weight determination subunit configured to assign a weight according to a time sequence of the effective detection data of the plurality of vehicles in a specific time window to obtain a time weight; a space weight determination subunit configured to assign a weight according to a distance between a vehicle and a standard detection point of a streetlight to obtain a space weight; and a credibility weight determination subunit configured to assign a weight according to a historical detection accuracy rate of a vehicle to obtain a credibility weight.

[0191] In an embodiment, the acquisition unit 302 is further configured to automatically adjust a sliding window length according to a traffic flow density, and adjust a sampling frequency of a region in combination with a key road section.

[0192] In an embodiment, the fusion unit 306 comprises: a weighted summation subunit configured to perform a weighted summation on the effective detection data of each vehicle in combination with the weight to obtain a weighted detection value; a mean value calculation subunit configured to calculate a mean value of the weighted detection values of all vehicles to obtain a streetlight instantaneous state value; and a merging subunit configured to merge the streetlight instantaneous state value, a historical trend analysis value, and a plurality of detection consistency evaluation values in combination with respective weight coefficients to obtain a comprehensive illumination value.

[0193] In an embodiment, the comparative analysis unit 307 comprises: a threshold value determination subunit configured to determine a judgment threshold value according to the historical data and a current environmental factor; wherein the judgment threshold value is dynamically adjusted by combining a weighted sum of a mean value and a standard deviation calculated from the historical data and the environmental impact factor, and is continuously updated by using a sliding window mechanism to adapt to environmental changes; and a state determination subunit configured to determine a state of the streetlight according to a comparison between the comprehensive illumination value and the judgment threshold value to update a streetlight state archive.

[0194] It should be noted that the specific implementation process of the multi-terminal cooperative road illumination comprehensive evaluation system 300 and each unit can be clearly understood by those skilled in the art, and can refer to the corresponding description in the foregoing method embodiments. For the convenience and brevity of description, it will not be repeated here.

[0195] The multi-terminal cooperative road illumination comprehensive evaluation system 300 described above can be implemented in the form of a computer program, which can run on a computer device as shown in the figure. Figure 10

[0196] Please refer to Figure 10 , Figure 10 is a schematic block diagram of a computer device provided by an embodiment of the present application. The computer device 500 can be a server, wherein the server can be a stand-alone server or a server cluster composed of multiple servers.

[0197] Referring to Figure 10 , the computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a system bus 501, wherein the memory can include a non-volatile storage medium 503 and an internal memory 504.

[0198] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions which, when executed, can cause the processor 502 to perform a multi-terminal cooperative road illumination comprehensive evaluation method.

[0199] The processor 502 is configured to provide computing and control capabilities to support the operation of the entire computer device 500.

[0200] The internal memory 504 provides an environment for the running of the computer program 5032 in the non-volatile storage medium 503, which, when executed by the processor 502, can cause the processor 502 to perform a multi-terminal cooperative road illumination comprehensive evaluation method.

[0201] The network interface 505 is configured to perform network communication with other devices. Those skilled in the art can understand that Figure 10 the structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device 500 to which the scheme of the present application is applied. The specific computer device 500 can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0202] The processor 502 is configured to run the computer program 5032 stored in the memory to implement all steps of the multi-terminal cooperative road illumination comprehensive evaluation method.​

[0203] It should be understood that, in the embodiments of the present application, the processor 502 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0204] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the above-mentioned embodiments.

[0205] Therefore, the present application also provides a storage medium. The storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein the computer program is executed by a processor to make the processor execute all steps of the multi-terminal cooperative road illumination comprehensive evaluation method.

[0206] The storage medium can be a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and various computer-readable storage media that can store program codes.

[0207] Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in the above description in general terms. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0208] In several embodiments provided by the present application, it should be understood that the disclosed system and method can be implemented in other manners. For example, the division of the system embodiments is merely illustrative. For example, the division of the units can be divided in another manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. The disclosed embodiments of the present application are not limited to the described division.

[0209] The steps in the method embodiments of the present application can be adjusted, combined and deleted in sequence according to actual needs. The units in the system embodiments of the present application can be combined, divided and deleted according to actual needs. In addition, each functional unit in each embodiment of the present application can be integrated in a processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0210] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part of the prior art that makes a contribution, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a terminal or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application.

[0211] The above description is merely a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A multi-terminal collaborative method for comprehensive road illumination assessment, characterized in that, include: Construct a geographic coordinate database for streetlights; The system acquires street light illuminance values, image grayscale features, and timestamps collected from detection devices on different vehicles to obtain multi-vehicle detection data. Interference spectra in the detection data of the multiple vehicles are identified and removed to obtain interference-free detection data of the multiple vehicles. The detection data of the multi-vehicles after removing interference is verified based on the correlation coefficient between the street light illuminance value and the average gray value of the image, as well as the relative coefficient of the spectrum corresponding to the street light illuminance value, in order to determine the effective detection data of the multi-vehicles. The weights of the effective detection data for the multiple vehicles are determined based on the street light geographic coordinate database. The comprehensive illuminance value is determined by combining the effective detection data of the multiple vehicles with the corresponding weights. Based on the comparative analysis of the comprehensive illuminance value and historical data, the status of the streetlights is determined in order to update the streetlight status file.

2. The multi-terminal collaborative road illumination comprehensive evaluation method according to claim 1, characterized in that, The process of identifying and removing interference spectra from the detection data of the multiple vehicles to obtain interference-free detection data of the multiple vehicles includes: The spectral similarity of the detection data of the multiple vehicles is calculated. The detection data corresponding to the similarity that meets the requirements is confirmed as the real signal, and the detection data corresponding to the similarity that does not meet the requirements is confirmed as the interference signal. The interference signal is removed to obtain the multi-vehicle detection data with interference removed.

3. The multi-terminal collaborative road illumination comprehensive evaluation method according to claim 1, characterized in that, The method of verifying the interference-removed multi-vehicle detection data based on the correlation coefficient between streetlight illuminance values ​​and average image grayscale values, as well as the spectral relative coefficients corresponding to streetlight illuminance values, to determine the effective detection data for multiple vehicles includes: The effectiveness of the multi-vehicle detection data after removing interference is verified based on the correlation coefficient between the street light illuminance value and the average gray value of the image, so as to determine the first effective data; The spectral relative coefficients corresponding to the street light illuminance values ​​are calculated based on the first valid data to determine the valid detection data for multiple vehicles.

4. The multi-terminal collaborative road illumination comprehensive evaluation method according to claim 3, characterized in that, The effectiveness of the interference-removed multi-vehicle detection data is verified based on the correlation coefficient between street light illuminance values ​​and average image grayscale values ​​to determine the first valid data, including: For each vehicle, the correlation coefficient between the street light illuminance value and the average gray value of the image is calculated based on the ratio of covariance to standard deviation of the detection data to measure the internal consistency of the detection data for a single vehicle. For multiple vehicles, within the same time period, the correlation coefficient between the street light illuminance value and the average gray value of the image is calculated based on the ratio of covariance to standard deviation of the detection data, and the correlation coefficients calculated for each vehicle are compared to measure the consistency and reliability of the detection data for multiple vehicles and determine the first valid data. Specifically, when the correlation coefficient of a vehicle deviates from the correlation coefficients of other vehicles, the weight of that vehicle is reduced.

5. The multi-terminal collaborative road illumination comprehensive evaluation method according to claim 3, characterized in that, The step of calculating the spectral relative coefficient corresponding to the street light illuminance value based on the first valid data to determine the valid detection data for multiple vehicles includes: For the first valid data, calculate the product of the spectral response and the exponentially decaying interference intensity function within the current time window to obtain the spectral relative coefficient corresponding to the street light illuminance value; The relationship between the spectral relative coefficient corresponding to the street light illuminance value and a set threshold is used to determine the effective detection data for multiple vehicles.

6. The multi-terminal collaborative road illumination comprehensive evaluation method according to claim 1, characterized in that, The step of determining the weights of the effective detection data for the multiple vehicles based on the street light geographic coordinate database includes: When different vehicles pass through the same location, the detection data of the multiple vehicles are automatically identified and associated based on the street light geographic coordinate database and a GPS matching algorithm. Within a specific time window, weights are assigned based on the temporal order of the valid detection data of the multiple vehicles to obtain time weights; The spatial weights are obtained by assigning weights based on the distance between the vehicle and the standard street light detection point; The reliability weight is obtained by assigning weights based on the historical detection accuracy of the vehicle.

7. The multi-terminal collaborative road illumination comprehensive evaluation method according to claim 1, characterized in that, The process of acquiring streetlight illuminance values, image grayscale features, and timestamps collected from detection devices on different vehicles to obtain multi-vehicle detection data includes: The sliding window length is automatically adjusted based on traffic density, and the sampling frequency of key road sections is adjusted accordingly.

8. The multi-terminal collaborative road illumination comprehensive evaluation method according to claim 1, characterized in that, The determination of the comprehensive illuminance value by combining the effective detection data of the multiple vehicles with corresponding weights includes: The valid detection data for each vehicle are weighted and summed according to the aforementioned weights to obtain a weighted detection value; The instantaneous state value of the street light is obtained by averaging the weighted detection values ​​of all vehicles. The instantaneous state value, historical trend analysis value, and consistency evaluation value of the street light are combined according to their respective weighting coefficients to obtain the comprehensive illuminance value.

9. The multi-terminal collaborative road illumination comprehensive evaluation method according to claim 1, characterized in that, The step of determining the street light status based on the comparative analysis of the comprehensive illuminance value and historical data, and updating the street light status file, includes: The judgment threshold is determined based on the historical data and current environmental factors. The threshold is dynamically adjusted by combining the average and standard deviation of historical data with the weighted sum of environmental impact factors, and continuously updated using a sliding window mechanism to adapt to environmental changes. The status of the streetlights is determined by comparing the comprehensive illuminance value with the judgment threshold, so as to update the streetlight status file.

10. A multi-terminal collaborative road illumination comprehensive assessment system, characterized in that, include: Building unit, used to build street light geographic coordinate database; The acquisition unit is used to acquire street light illuminance values, image grayscale features, and timestamps collected from detection devices on different vehicles to obtain multi-vehicle detection data. An identification and elimination unit is used to identify and eliminate interference spectra in the detection data of the multiple vehicles to obtain interference-free detection data of the multiple vehicles. The validity determination unit is used to verify the interference-removed multi-vehicle detection data based on the correlation coefficient between the street light illuminance value and the average gray value of the image and the spectral relative coefficient corresponding to the street light illuminance value, so as to determine the valid detection data of the multi-vehicle. A weight determination unit is used to determine the weights of the effective detection data of the multiple vehicles based on the street light geographic coordinate database. The fusion unit is used to determine the comprehensive illuminance value by combining the effective detection data of the multiple vehicles with corresponding weights. The comparison and analysis unit is used to determine the status of the streetlights based on the comparison and analysis of the comprehensive illuminance value and historical data, so as to update the streetlight status file.

Images