Online passenger flow monitoring analysis data error correction method

Abstract

The invention provides an online passenger flow monitoring analysis data error correction method, which comprises the following steps of: setting a virtual path in a monitoring area, segmenting the virtual path according to a set length interval to form a plurality of monitoring sections, setting a plurality of monitoring points on the virtual path of each monitoring section, face recognition units are arranged at the starting point and the end point of the virtual path in the monitoring section; data association is carried out on a plurality of monitoring sections, each monitoring section carries out section statistics on monitoring data of the monitoring section and then uniformly uploads the monitoring data to a data analysis unit, and a personnel information database is reserved in the data analysis unit to store the face information uploaded by the face recognition unit; a data elimination unit is arranged at the end point of the virtual path, when a user passes through a set end point section, stored face information is called and then matched, and after matching is completed, the face information is deleted, and then the matched information is stored in a data analysis unit.

Description

Technical Field

[0001] The invention belongs to the field of passenger flow monitoring, and in particular relates to an online passenger flow monitoring and analysis data error correction method. Background Art

[0002] Online passenger flow monitoring is a method of real-time monitoring and analysis of passenger flow using sensors, video surveillance and other technical means. By monitoring and analyzing passenger flow, it can help businesses, urban planners, etc. make decisions, optimize passenger flow layout, and improve service quality.

[0003] At present, there are some online passenger flow monitoring and analysis systems that can collect and analyze passenger flow data, but due to various factors, these data may have certain errors. Therefore, in order to improve the accuracy and reliability of the data, error correction is required.

[0004] A common error correction method is a sensor-based online passenger flow monitoring system. This system senses the flow of people by installing sensor devices at specific locations, such as entrances or exits. Then, by analyzing and processing the sensor data, an estimate of the passenger flow can be obtained. However, due to factors such as the accuracy limitation of the sensor itself and environmental interference, the sensor data may have certain errors.

[0005] In order to correct these errors, one approach is to introduce other auxiliary data sources. For example, video surveillance technology can be combined to analyze video images, detect and track crowds, and then obtain an estimate of passenger flow. Compared with sensors, video surveillance has higher precision and accuracy and can better solve the problem of sensor errors.

[0006] In addition, machine learning and other technologies can be used to correct passenger flow data. By building a data model and training and learning sensor data, data errors can be discovered and corrected. For example, historical data and real-time data can be used to train models to predict and correct passenger flow data errors.

[0007] In summary, the existing online passenger flow monitoring and analysis data error correction methods mainly include sensor technology, video surveillance technology, and machine learning technology. These methods can be combined with each other to improve the accuracy and reliability of data from different angles. Summary of the invention

[0008] The present invention proposes an online passenger flow monitoring and analysis data error correction method, which aims to solve the problem of inaccurate data caused by face recognition unit errors and environmental factors in online passenger flow monitoring and analysis. By setting virtual paths and monitoring points, using face recognition units and data elimination units, combining integration and segmented analysis in the data analysis unit, and collecting and correcting traffic efficiency data, the real-time accuracy and reliability of passenger flow monitoring data are improved.

[0009] The technical solution of the present invention is implemented as follows: a method for correcting errors in online passenger flow monitoring and analysis data, the method comprising the following steps:

[0010] A virtual path is set in the monitoring area, and the virtual path is segmented according to a set length interval to form a number of monitoring segments, a number of monitoring points are set on the virtual path of each monitoring segment, and a face recognition unit is set at the starting point and the end point of the virtual path in the monitoring segment;

[0011] Data of several monitoring segments are associated, and each monitoring segment performs segment statistics on its own monitoring data and uploads them to the data analysis unit. A personnel information database is reserved in the data analysis unit to store the face information uploaded by the face recognition unit;

[0012] A data deletion unit is set at the end point of the virtual path. When the user passes through the set end point section, the stored face information is retrieved for matching. After the matching is completed, the face information is deleted and the matching information is retained in the data analysis unit.

[0013] The data analysis unit integrates the data in each monitoring section, performs segmented analysis on the integrated data, collects the paths of passenger flow in the monitoring area and sorts them by time, and counts the average time data of passenger flow passing through each monitoring section. After several repeated collections, it obtains the traffic efficiency data under different weather and time conditions, and makes real-time corrections to the number of people entering the monitoring area based on the collected traffic efficiency data.

[0014] Compared with the existing technology, the main differences of this method include: Virtual path setting and segmentation: This method first sets up virtual paths in the monitoring area and divides these paths into segments. This method can more accurately monitor and analyze the flow of passenger flow through the setting of virtual paths and monitoring points, rather than just simple import and export statistics.

[0015] Application of face recognition unit: Face recognition units are set at the start and end points of each monitoring section, which allows the system to track the passage of individuals by identifying repeated faces. This is different from the existing technology that may only use sensors or video surveillance, and increases the accuracy of face recognition.

[0016] Data association and upload: The data of the monitoring section is uniformly uploaded to the data analysis unit after segment statistics, which helps to centrally process and analyze the data, while the reserved personnel information database is used to store facial information for subsequent matching and correction.

[0017] Introduction of data elimination unit: A data elimination unit is set at the end of the virtual path to eliminate the matched face information when the user passes through the end section to avoid repeated counting. This is a rare step in the prior art and helps to improve the accuracy of the data.

[0018] Collection and correction of traffic efficiency data: This method collects the average time data of people passing through each monitoring section and collects traffic efficiency data under different weather and time conditions to correct the number of people entering the monitoring area in real time. This method combines data analysis of actual traffic conditions, which is different from the existing methods that may only rely on real-time data or historical data for correction, and is more comprehensive and accurate.

[0019] This method provides a more sophisticated and accurate online passenger flow monitoring and analysis data error correction solution by introducing virtual paths, face recognition, data elimination units, and the collection and analysis of traffic efficiency data. Compared with the existing technology, it has improvements and innovations in data accuracy, personal privacy protection, and real-time correction.

[0020] As a preferred embodiment, when setting up a virtual path, the entrances, exits and intermediate points of all channels are connected in series, and the channels are interconnected to form a virtual path. The traffic heat map of the virtual path is obtained through the traffic efficiency data of the data analysis unit, and the traffic heat map of the virtual path is segmented according to the monitoring sections and sent to the visualization platform for display.

[0021] As a preferred implementation, when data is associated with several monitoring segments, a synchronization association threshold is set. The synchronization association threshold is the data synchronization error in the monitoring segment. Adjacent monitoring segments first perform data mutual verification. When the synchronization association threshold of adjacent monitoring segments is within the set threshold, the data association is maintained. When the synchronization association threshold of adjacent monitoring segments exceeds the set threshold, an alarm is issued and the data is recorded and reset before association is performed.

[0022] As a preferred embodiment, the face recognition unit first calibrates the face position while monitoring the image of the monitoring area. After the calibration is completed, the face image is collected according to the calibrated position. If the face position is not within the calibration area, the face information is not collected. If the face position is within the calibration area, data image collection is performed by identifying the tip of the nose as the center point.

[0023] As a preferred implementation, when the data analysis unit integrates the data in each monitoring segment, it first unifies the data time in the monitoring segment, and establishes the dynamic trend of human flow in sequence according to the monitoring segment sequence. When performing segmented analysis on the integrated data, the dynamic trend of human flow is used as the main parameter, and the unrecognized facial information is used as the error parameter. After the error parameters of each monitoring segment are sorted and normalized in the last monitoring segment, the main parameter is corrected.

[0024] After adopting the above technical scheme, the beneficial effect of the present invention is that by setting virtual paths and monitoring points, combined with face recognition technology, the flow of passenger flow can be tracked and recorded more accurately, thereby improving the accuracy of data. The data elimination unit is introduced to eliminate the matched face information when the user passes through the terminal section, effectively avoiding the problem of repeated data counting caused by the error of the face recognition unit or the repeated passage of the user in the monitoring area. In the data processing process, once the face information is matched with the user and used for statistical analysis, it is deleted, which helps to protect the user's personal privacy. By collecting and analyzing the traffic efficiency data under different weather and time conditions, the system can correct the number of people entering the monitoring area in real time, providing merchants and urban planners with instant and accurate data support to help them make quick decisions. By sorting and segmenting the path collection of passenger flow in the monitoring area, the distribution and behavior pattern of passenger flow can be more deeply understood, providing more insights for future passenger flow management and optimization. This method not only improves the accuracy of online passenger flow monitoring data, but also takes into account personal privacy protection. At the same time, through real-time correction and in-depth analysis, it provides a more effective and practical tool for passenger flow management and optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0026] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Example:

[0029] like Figure 1 As shown, a method for correcting errors in online passenger flow monitoring and analysis data includes the following steps:

[0030] A virtual path is set in the monitoring area, and the virtual path is segmented according to a set length interval to form a number of monitoring segments, a number of monitoring points are set on the virtual path of each monitoring segment, and a face recognition unit is set at the starting point and the end point of the virtual path in the monitoring segment;

[0031] Data of several monitoring segments are associated, and each monitoring segment performs segment statistics on its own monitoring data and uploads them to the data analysis unit. A personnel information database is reserved in the data analysis unit to store the face information uploaded by the face recognition unit;

[0032] A data deletion unit is set at the end point of the virtual path. When the user passes through the set end point section, the stored face information is retrieved for matching. After the matching is completed, the face information is deleted and the matching information is retained in the data analysis unit.

[0033] The data analysis unit integrates the data in each monitoring section, performs segmented analysis on the integrated data, collects the paths of passenger flow in the monitoring area and sorts them by time, and counts the average time data of passenger flow passing through each monitoring section. After several repeated collections, it obtains the traffic efficiency data under different weather and time conditions, and makes real-time corrections to the number of people entering the monitoring area based on the collected traffic efficiency data.

[0034] The workflow and working principle of the online passenger flow monitoring and analysis data error correction method are as follows: In the monitoring area, a virtual path is first set up and divided into segments according to the set length intervals. Several monitoring points are set on the virtual path of each monitoring segment, and these monitoring points are used to record the entry and exit of passenger flow. A face recognition unit is set at the starting point and the end point of the virtual path of each monitoring segment. When a user passes through the monitoring area, the face recognition unit will identify the user's facial information, match it with the personnel information database and record it. Each monitoring segment will perform segment statistics on its own monitoring data and upload it to the data analysis unit in a unified manner. The personnel information database reserved in the data analysis unit is used to store the facial information uploaded by the face recognition unit for subsequent matching and correction.

[0035] A data elimination unit is set at the end of the virtual path. When a user passes through the set end section, the data elimination unit will retrieve the stored facial information and match it with the current user. After the matching is completed, the matched facial information will be deleted, leaving only the matching information in the data analysis unit to avoid repeated counting. The data analysis unit integrates the data of each monitoring section and performs segmented analysis. By sorting the path collection of passenger flow in the monitoring area in time, the average time data of passenger flow passing through each monitoring section is counted to understand the distribution and behavior pattern of passenger flow. By repeatedly collecting traffic efficiency data under different weather and time conditions, the system can correct the number of people entering the monitoring area in real time and provide accurate data support. This can help merchants and urban planners make immediate decisions and optimize passenger flow management and services. Through the setting and segmentation of virtual paths, the application of face recognition units, the introduction of data elimination units, and the collection and correction of traffic efficiency data. These differences enable this method to more accurately monitor and analyze the flow of passenger flow, avoid repeated counting, and provide real-time correction and optimization capabilities, thereby improving the accuracy of data and the depth of analysis.

[0036] As a preferred embodiment, when setting a virtual path, the entrances, exits and intermediate points of all channels are connected in series, and the channels are connected to each other to form a virtual path, and the traffic heat map of the virtual path is obtained through the traffic efficiency data of the data analysis unit, and the traffic heat map of the virtual path is segmented according to the monitoring section and sent to the visualization platform for display. In this method, the virtual path is not just a simple connection of the entrances, exits and intermediate points of the channel, but also includes connecting these channels to each other to form a complete path network. This construction method helps to more accurately simulate and reflect the actual passenger flow, because it takes into account the connectivity between different channels and the flow path of passenger flow. Through the traffic efficiency data of the data analysis unit, the method can generate a traffic heat map of the virtual path. This heat map can intuitively show the passenger flow density and flow conditions of different monitoring sections, providing merchants and urban planners with an effective tool to analyze and understand passenger flow distribution. Different from the overall traffic heat map that may be provided in the prior art, this method segments the traffic heat map of the virtual path according to the monitoring section, so that detailed passenger flow information can be provided for each monitoring section. This segmented display method helps to analyze and monitor passenger flow more accurately, especially when passenger flow is unevenly distributed or there are multiple hot spots. The generated traffic heat map is sent to the visualization platform for display, which provides users with an intuitive interface to view and analyze passenger flow data. Compared with the existing technology that may only provide data reports or tables, the visualization platform can display data in a graphical way, allowing users to understand and interpret passenger flow more intuitively. This method provides a more sophisticated and intuitive online passenger flow monitoring and analysis method by constructing a complete virtual path, generating a traffic heat map, segmented display, and using a visualization platform. Compared with the existing technology, it has improvements and innovations in the depth of data analysis, the intuitiveness of visualization, and the accuracy of passenger flow monitoring.

[0037] When several monitoring segments are associated with data, a synchronization association threshold is set. The synchronization association threshold is the data synchronization error in the monitoring segment. Adjacent monitoring segments first perform data mutual verification. When the synchronization association threshold of the adjacent monitoring segments is within the set threshold, the data association is maintained. When the synchronization association threshold of the adjacent monitoring segments exceeds the set threshold, an alarm is issued, and the data is recorded and reset before association is performed. The method introduces a synchronization association threshold to judge the data synchronization error of adjacent monitoring segments. When the synchronization association threshold of the adjacent monitoring segments is within the set threshold, the data association is maintained; when it exceeds the set threshold, an alarm is issued, and the data is recorded and reset before association is performed. By setting a reasonable synchronization association threshold, the impact of data error on the accuracy of associated data can be reduced. In this method, the data of adjacent monitoring segments are mutually verified to determine whether their synchronization association thresholds are within the set threshold. This data mutual verification method helps to discover and correct data anomalies and improve the accuracy and reliability of data association. When the synchronization association threshold of the adjacent monitoring segments exceeds the set threshold, an alarm is issued, and the data is recorded and reset before association is performed. This alarm and data recording method can timely discover data anomalies and take corresponding measures to deal with them. At the same time, recording abnormal data also helps with subsequent data analysis and problem tracking. This method provides a more reliable and accurate data association method by setting synchronous association thresholds, data mutual verification of adjacent monitoring segments, alarms, and data recording. Compared with the existing technology, it has improved and innovated in the accuracy and reliability of data association, can better handle data errors and abnormal situations, and improve the credibility of data association results.

[0038] While the face recognition unit monitors the image of the monitoring area, it first calibrates the face position. After the calibration is completed, the face image is collected according to the calibrated position. If the face position is not within the calibration area, the face information is not collected. If the face position is within the calibration area, the data image is collected by identifying the nose tip as the center point. The method first calibrates the face position. This means that before starting to collect the face image, the system will determine the specific position of the face in the image of the monitoring area. The introduction of this step can improve the accuracy and efficiency of face recognition because it ensures that only the face in the calibration area will be collected and recognized. After the calibration is completed, the system collects the face image according to the calibrated position. This image acquisition method based on the calibration position helps to improve the quality of the face image because the collected image is more focused on the face area, thereby reducing the interference of background noise and non-face areas. If the face position is within the calibration area, the system collects the data image by identifying the nose tip as the center point. This strategy can improve the accuracy of face recognition because the nose tip is usually a part where the facial features are more obvious and stable. By taking the nose tip as the center point for image acquisition, the key features of the face can be better captured, thereby improving the recognition accuracy. This method provides a new face recognition method that is more accurate and has a higher face recognition accuracy by introducing face position calibration, image acquisition based on the calibration position, and a recognition strategy using the nose tip as the center point. Compared with the existing technology, it has improvements and innovations in face recognition accuracy and image acquisition strategies.

[0039] When the data analysis unit integrates the data in each monitoring segment, it first unifies the data time in the monitoring segment, establishes the dynamic trend of the flow of people in sequence according to the order of the monitoring segments, and when the integrated data is segmented and analyzed, the dynamic trend of the flow of people is used as the main parameter, and the unrecognized face information is used as the error parameter. After the error parameters of each monitoring segment are sorted and normalized in the last monitoring segment, the main parameter is corrected. A method for the data analysis unit to integrate and analyze the monitoring segment data includes unified processing of data time, establishment of dynamic trend of flow of people and use of error parameters. Compared with the prior art, the main difference is that before integrating the data, the data analysis unit first unifies the data time in each monitoring segment. This means that the system adjusts the data time in different monitoring segments so that they are on the same time basis. This ensures that the data is accurate and consistent in time, providing an accurate time reference for subsequent analysis. According to the order of the monitoring segments, the data analysis unit sequentially establishes the dynamic trend of the flow of people. The establishment of this dynamic trend helps to capture and understand the flow pattern and behavior of passenger flow, and provides important reference information for subsequent segment analysis and optimization. During the segmented analysis, the data analysis unit uses the dynamic trend of passenger flow as the main parameter. This means that the system mainly relies on the flow trend of passenger flow for analysis and decision-making, which helps to understand the distribution and behavior of passenger flow more accurately. During the analysis process, unrecognized facial information is used as an error parameter. This means that the system regards unrecognized facial information as a possible source of error and uses it to correct the main parameters. This practice can improve the accuracy of data analysis because it takes into account possible data errors and uncertainties. After the error parameters of each monitoring segment are sorted and normalized in the last monitoring segment, the main parameters are corrected. This normalization process helps to standardize the error parameters in different monitoring segments, so that the main parameters can be corrected more accurately, improving the accuracy and reliability of data analysis.

[0040] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for correcting online passenger flow monitoring and analysis data errors, characterized in that: The method comprises the following steps: A virtual path is set in the monitoring area, and the virtual path is segmented according to a set length interval to form a number of monitoring segments, a number of monitoring points are set on the virtual path of each monitoring segment, and a face recognition unit is set at the starting point and the end point of the virtual path in the monitoring segment; Data of several monitoring segments are associated, and each monitoring segment performs segment statistics on its own monitoring data and uploads them to the data analysis unit. A personnel information database is reserved in the data analysis unit to store the face information uploaded by the face recognition unit; A data deletion unit is set at the end point of the virtual path. When the user passes through the set end point section, the stored face information is retrieved for matching. After the matching is completed, the face information is deleted and the matching information is retained in the data analysis unit. The data analysis unit integrates the data in each monitoring section, performs segmented analysis on the integrated data, collects the paths of passenger flow in the monitoring area and sorts them by time, and counts the average time data of passenger flow passing through each monitoring section. After several repeated collections, it obtains the traffic efficiency data under different weather and time conditions, and makes real-time corrections to the number of people entering the monitoring area based on the collected traffic efficiency data.

2. The method for correcting online passenger flow monitoring and analysis data errors according to claim 1, characterized in that: When setting up a virtual path, the entrances, exits and intermediate points of all channels are connected in series, and the channels are interconnected to form a virtual path. The traffic heat map of the virtual path is obtained through the traffic efficiency data of the data analysis unit. The traffic heat map of the virtual path is segmented according to the monitoring sections and sent to the visualization platform for display.

3. The method for correcting online passenger flow monitoring and analysis data errors according to claim 1, characterized in that: When data is associated with several monitoring segments, a synchronization association threshold is set. The synchronization association threshold is the data synchronization error in the monitoring segment. Adjacent monitoring segments first perform data mutual verification. When the synchronization association threshold of adjacent monitoring segments is within the set threshold, the data association is maintained. When the synchronization association threshold of adjacent monitoring segments exceeds the set threshold, an alarm is issued and the data is recorded and reset before association is performed.

4. The method for correcting online passenger flow monitoring and analysis data errors according to claim 1, characterized in that: While the face recognition unit monitors the image of the monitoring area, it first calibrates the face position. After the calibration is completed, the face image is collected according to the calibrated position. If the face position is not within the calibration area, the face information is not collected. If the face position is within the calibration area, data image collection is performed by identifying the tip of the nose as the center point.

5. The method for correcting online passenger flow monitoring and analysis data errors according to claim 1, characterized in that: When the data analysis unit integrates the data in each monitoring segment, it first unifies the data time in the monitoring segment, and establishes the dynamic trend of human flow in sequence according to the monitoring segment sequence. When performing segmented analysis on the integrated data, the dynamic trend of human flow is used as the main parameter, and the unrecognized facial information is used as the error parameter. After the error parameters of each monitoring segment are sorted and normalized in the final monitoring segment, the main parameter is corrected.

Images