Intelligent logistics monitoring and early warning system based on cloud computing platform

Abstract

This invention provides an intelligent logistics monitoring and early warning system based on a cloud computing platform, relating to the field of logistics monitoring technology. It includes a sensor module, a data acquisition module, a data preprocessing module, a communication module, a server, a terminal module, and a notification module. This invention comprehensively assesses the driver's health status based on multi-dimensional health indicators such as the number of blood vessels in the eyeballs, the duration of eye closure intervals, and the amplitude of eye closure fluctuations. It fuses and analyzes the collected health status data using a multivariate nonlinear regression model to construct a health status change value H, reflecting the dynamic changes in the driver's health status. Based on the output value of the health status change value H, it outputs multi-level fatigue assessment information, integrates data from multiple modules, and outputs comprehensive and accurate early warning information to ensure the safety and efficiency of logistics drivers. Through multi-dimensional analysis and dynamic monitoring of health status data, it can significantly improve the accuracy and response speed of early warnings.

Description

Technical Field

[0001] This invention relates to the field of logistics monitoring technology, specifically to an intelligent logistics monitoring and early warning system based on a cloud computing platform. Background Technology

[0002] With the rapid development of the logistics industry, the transportation tasks of logistics vehicles are becoming increasingly heavy. Drivers work long hours and at high intensity, which can easily lead to health problems such as fatigue, insomnia, cervical spondylosis, and eye strain. These health problems not only affect the drivers' work efficiency but may also lead to traffic accidents, threatening road traffic safety.

[0003] Existing technologies for fatigue detection include the following:

[0004] Detection methods based on motor vehicle behavior characteristics: When a driver is fatigued, it will be reflected to some extent through the behavior of the vehicle he is operating. Vehicle behavior characteristic parameters can be easily obtained and used for fatigue detection, such as the pressure on the steering wheel from the driver's hands, changes in vehicle acceleration, vehicle sway, and the degree to which the vehicle deviates from the driving lane.

[0005] Detection methods based on driver physiological characteristics, which mainly refer to electrocardiogram signals, electroencephalogram signals, and body surface temperature, are highly accurate. However, drivers need to wear related physiological indicator detection equipment, which will affect the comfort of the driving process and cause inconvenience to the driver.

[0006] The driver's facial feature-based detection method refers to using driver facial images acquired through image sensors such as cameras as a basis, and employing machine vision algorithms such as face detection and facial feature point localization to extract and analyze facial features such as eye opening and closing, mouth opening and closing, and head posture, thereby enabling the analysis and judgment of the driver's fatigue state.

[0007] However, most existing technologies can only monitor a single indicator (such as heart rate or blood pressure), lacking a comprehensive assessment of multi-dimensional health status. Furthermore, current technologies struggle to dynamically adjust monitoring parameters based on actual transportation conditions, failing to accurately reflect the driver's true health status. Therefore, existing technologies still have significant shortcomings in terms of monitoring accuracy, early warning efficiency, and applicability.

[0008] The existing technology, with publication number CN119169596A, entitled "A Method for Detecting Fatigue Driving by Logistics Transportation Drivers," involves capturing driver images in real-time during driving. When both eye and mouth features are simultaneously detected, the number of times the driver's eyes close and mouth open is counted within a T-cycle. If the number of times the driver closes their eyes or opens their mouth reaches a preset number within a cycle, the driver is considered fatigued. When only eye features are detected and not mouth features, the eye area image is cropped, and a facial expression image is automatically matched to the eye area image based on the eye expression. Then, based on the facial shape change area in the matched facial expression image, the area in the driver's real-time facial image that may cause changes in the shape of the mask is determined, i.e., the detection area. Finally, based on the change in the shape of the mask stripes within the detection area, it is determined whether the driver is yawning; the driver is considered fatigued.

[0009] The information disclosed in the background section above is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0010] The purpose of this invention is to provide an intelligent logistics monitoring and early warning system based on a cloud computing platform to solve the problems mentioned in the background art.

[0011] To achieve the above objectives, the present invention provides the following technical solution:

[0012] The intelligent logistics monitoring and early warning system based on a cloud computing platform includes a sensor module, a data acquisition module, a data preprocessing module, a communication module, a server, a terminal module, and a notification module.

[0013] The sensor module is used to monitor logistics drivers, and the collected data includes taking photos of the eyeballs, the duration of the eye-closing interval, and the amplitude of the eye-closing fluctuation.

[0014] The communication module is used to upload the data collected by the sensor module and to transmit data, communication and instructions for the entire system.

[0015] The server is used to receive data transmitted by the communication module and to analyze and process the data through the model inside the server.

[0016] The process involves capturing and processing photographs of the eyeballs to calculate blood vessel density: Preprocessing the eyeball photographs flattens them to maintain their planarity; image editing software is used to crop the planar photographs, removing the black part of the eye; the cropped photographs retain only the white part of the eye, clearly displaying blood vessel traces; blood vessel traces are identified in the white part of the eyeball based on color channel detection; the identification results include the number and region of blood vessels; the density of blood vessels within each region is calculated; and the analysis structures of the number of blood vessels in the eyeball, the duration of eye closure intervals, and the amplitude of eye closure fluctuations are fused to construct a joint analysis of health status change values, comprehensively assessing the driver's health status, and generating fatigue assessment information based on these health status change values.

[0017] The formula for calculating the change in health status is as follows:

[0018] ;

[0019] in, Represented as a change in health status. This represents the duration of eye closure, and This is represented as a sequence of positions in the dataset of eye-closing intervals. This represents the minimum duration of the eye-closing interval. This is expressed as the amplitude of fluctuation when eyes are closed, and This is represented as the position sequence in the closed-eye fluctuation amplitude dataset. This represents the maximum amplitude of the fluctuation when the eyes are closed. Expressed as blood vessel density, , , and All are represented as model parameters;

[0020] The terminal module is used to integrate data from multiple modules, output comprehensive and accurate early warning information based on fatigue assessment information, and construct early warning levels.

[0021] Furthermore, the sensor module includes an eye imager of an eye imaging device and an eye tracker of an eye movement monitoring device;

[0022] The sensor module is located at the front of the logistics vehicle and above the driver's position, enabling it to take photos of the driver's eyes and monitor the duration of eye closure intervals.

[0023] The eye imaging device is used to take pictures of the driver's eyes, and the eye movement monitoring device is used to monitor the duration of the driver's eye closure intervals and the amplitude of eye closure fluctuations.

[0024] Furthermore, the data acquisition module is integrated inside the sensor module, and the data acquisition module is used to control the eye imaging device and the eye tracker to take pictures of the eyeball and monitor the duration of the eye closure interval and the amplitude of the eye closure fluctuation.

[0025] The data preprocessing module is also integrated inside the sensor module. The data preprocessing module is used to preprocess the eye photos and monitoring intervals of eye closure acquired by the data acquisition module to obtain preprocessed data.

[0026] Furthermore, the server includes a cloud computing platform, which analyzes and processes the data through a model;

[0027] The cloud computing platform also includes an extraction module for extracting features from the data to obtain feature sequences, and a model module for analyzing and processing the feature sequences. The model module first analyzes and processes the feature sequences to obtain health status change values, and then the model module analyzes and processes the health status change values ​​to obtain fatigue assessment information.

[0028] Furthermore, the cloud computing platform performs the following steps to analyze and process the transmitted data:

[0029] The server receives and stores data: it receives and stores the data transmitted by the communication module, and stores the received data to the cloud computing platform to form a complete dataset of driver health status.

[0030] Data analysis and model building: Technical features were extracted from the driver health status dataset, and model modules were built for the number of blood vessels in the eyes, the duration of the eye closure interval, and the amplitude of fluctuations when the eyes are closed.

[0031] Health status data fusion and analysis: The driver health status dataset is fused and analyzed to obtain a joint analysis of the number of blood vessels in the eyeballs, the duration of the eye closure interval, and the fluctuation amplitude of the eye closure, and to obtain the health status change value H;

[0032] The model module calculates the change in health status: analyzes the range of numerical parameters of the change in health status H, and assesses the driver's health status.

[0033] Fatigue assessment generation: The model module is used to calculate the health status change value H. Based on the change of the health status change value H, the driver's fatigue level is assessed and fatigue assessment information is output.

[0034] Furthermore, the feature extraction steps for the driver health status dataset are as follows:

[0035] First, the preprocessed data of the eyeball image is planarized to maintain the planarity of the preprocessed eyeball image;

[0036] The image was cropped using image editing software to remove the black part of the eye; the cropped image only retains the white part of the eye, clearly showing the bloodshot marks.

[0037] Color channel-based blood vessel detection identifies blood vessels in the whites of the eyes; the results include the number and region of blood vessels; and the density of blood vessels within the region is calculated.

[0038] Furthermore, the analysis of the duration of the eye-closing interval is as follows:

[0039] Preprocessed data on the duration of eye closure intervals monitored by eye-tracking devices are subjected to time-series processing;

[0040] The closed-eye nodes are linked to the time series, and features are extracted from the closed-eye nodes in the time series.

[0041] The technical feature extraction for the eye-closing interval duration data includes the start and end values ​​of the interval duration time series, and then the eye-closing interval duration and the fluctuation amplitude of the eye-closing.

[0042] Furthermore, the formula for calculating the health status change value H is as follows:

[0043] ;

[0044] in, Represented as a change in health status. This represents the duration of eye closure, and This is represented as a sequence of positions in the dataset of eye-closing intervals. This represents the minimum duration of the eye-closing interval. This is expressed as the amplitude of fluctuation when eyes are closed, and This is represented as the position sequence in the closed-eye fluctuation amplitude dataset. This represents the maximum amplitude of the fluctuation when the eyes are closed. Expressed as blood vessel density, , , and All of these are represented as model parameters.

[0045] Furthermore, the health status change value H generates fatigue assessment information as follows:

[0046] ;

[0047] in, This represents a classification of fatigue levels. Represented as a change in health status, ranging from [0, 1]; when When the value approaches 0, it indicates a stable health status, corresponding to a low fatigue level; when... When the value approaches 1, it indicates an abnormal health condition or extreme fatigue, corresponding to a severe fatigue level.

[0048] Through changes in health status values The multi-level assessment comprehensively evaluates driver fatigue, providing a basis for generating early warning information.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] This invention uses an eye imaging device and an eye tracker to collect real-time eye photos, eye closure interval duration, and eye closure fluctuation amplitude, among other multi-dimensional health indicators, to comprehensively assess the driver's health status. Furthermore, by planarizing and cropping the eye photos to retain only the sclera, it facilitates the collection of blood vessel traces and divides the images into regions to calculate the density of blood vessels within each region. This allows for a comprehensive assessment of the driver's health status based on multi-dimensional health indicators such as the number of blood vessels in the eye, eye closure interval duration, and eye closure fluctuation amplitude.

[0051] This invention uses a multivariate nonlinear regression model to fuse and analyze the collected health status data, constructing a health status change value H to reflect the dynamic changes in the driver's health status; based on the output value of the health status change value H, multi-level fatigue assessment information is output to help drivers detect and deal with fatigue in a timely manner, and the monitoring parameters are dynamically adjusted according to the actual transportation situation to ensure the accuracy and applicability of the monitoring results.

[0052] This invention integrates data from multiple modules to output comprehensive and accurate early warning information, ensuring the safety and efficiency of logistics drivers. Through multi-dimensional analysis and dynamic monitoring of health status data, it can significantly improve the accuracy and response speed of early warnings, reduce traffic accidents caused by driver fatigue, and improve the overall safety and efficiency of logistics transportation. It is especially suitable for long-distance logistics or logistics tasks requiring long-duration driving. By monitoring and evaluating the driver's health status in real time, it can help the driver adjust their work status in a timely manner, reduce work fatigue, and improve transportation safety and efficiency. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the overall system structure of the present invention;

[0054] Figure 2 This is a schematic diagram illustrating the steps of the cloud computing platform of the present invention in analyzing and processing transmitted data information.

[0055] Figure 3 This is a schematic diagram illustrating the feature extraction steps of the driver health status dataset of the present invention;

[0056] Figure 4This is a schematic diagram illustrating the steps for analyzing the duration of eye-closing intervals in this invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0058] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0059] Example 1:

[0060] Please see Figure 1-4 This invention provides a technical solution: an intelligent logistics monitoring and early warning system based on a cloud computing platform, comprising a sensor module, a data acquisition module, a data preprocessing module, a communication module, a server, a terminal module, and a notification module.

[0061] The sensor module is used to monitor logistics drivers, and the collected data includes taking photos of the eyeballs, the duration of the eye-closing interval, and the amplitude of the eye-closing fluctuation.

[0062] In this embodiment, preferably, the sensor module includes an ocular imager of an ocular imaging device and an eye tracker of an eye movement monitoring device;

[0063] The sensor module is located at the front of the logistics vehicle and above the driver's position, enabling it to take photos of the driver's eyes and monitor the duration of eye closure intervals.

[0064] The eye imaging device is used to take pictures of the driver's eyeballs, and the eye movement monitoring device is used to monitor the driver's eye closure interval duration and eye closure fluctuation amplitude;

[0065] It should be noted that by taking photos of the driver's eyes with an eye imaging device and monitoring the duration and amplitude of the driver's eye closure intervals with eye movement monitoring equipment, the most basic parameter information can be obtained. The obtained parameter information is convenient for subsequent accurate calculation and analysis.

[0066] In this embodiment, preferably, the data acquisition module is integrated inside the sensor module. The data acquisition module is used to control the eye imaging device and the eye tracker to take pictures of the eyeballs and monitor the duration of the eye closure interval and the amplitude of the eye closure fluctuation.

[0067] The data preprocessing module is also integrated inside the sensor module. The data preprocessing module is used to preprocess the eye photos and monitoring the duration of eye closure intervals acquired by the data acquisition module to obtain preprocessed data.

[0068] It should be noted that the data acquisition module collects the corresponding data parameters detected by the sensor module, and preprocesses the collected eye photos and the monitoring interval of eye closure to improve the accuracy of the data parameters.

[0069] Use an ocular imaging device (such as a fundus camera or ocular imager) to take a picture of the driver's eyes;

[0070] Use image processing software to preprocess the photo, including brightness adjustment, contrast enhancement, etc.

[0071] Ensure image quality by removing noise and blurry parts; obtain clear images of the eyeballs to provide high-quality data for subsequent feature extraction.

[0072] Use an ocular imaging device to take a picture of the eye;

[0073] Image analysis software was used to extract the number of blood vessels and the duration of the eye-closing interval from the photos.

[0074] Output the extraction results, including specific values ​​for the number of blood vessels and the duration of the eye-closing interval;

[0075] Obtaining the number of blood vessels in a driver's eyes and the duration of eye closure provides important data for health status assessment.

[0076] Preprocessing of eyeball photos

[0077] Noise reduction: Use a Gaussian filter or median filter to denoise the eye photos to ensure image clarity;

[0078] Brightness Adjustment: Use the curve tool to adjust the brightness of the photo to ensure that the eye area is clearly visible;

[0079] Contrast Enhancement: Use the contrast enhancement tool to highlight the bloodshot marks in the white area of ​​the eyeball;

[0080] Color balance: Adjust the color balance to ensure consistent color tones in the photo and avoid color distortion;

[0081] Function: To ensure the clarity and quality of eyeball images, providing high-quality data for subsequent analysis.

[0082] Cropping of eyeball photos

[0083] Locate the center of the eyeball: Use image recognition algorithms or manual adjustments to determine the center position of the eyeball;

[0084] Cropping to remove black eyeballs: Based on the center of the eyeball, cropping to remove the black eyeball area outside the photo;

[0085] Preserve the whites of the eyes: After cropping, preserve the whites of the eyes to ensure that the bloodshot marks are clearly visible;

[0086] Function: Focuses on the white of the eye area, facilitating subsequent identification and analysis of blood vessels.

[0087] Identification of blood streaks

[0088] Bloodshot detection algorithm: A color channel-based bloodshot detection algorithm is used to identify bloodshot traces on the whites of the eyes;

[0089] Blood vessel count: The number of blood vessels is counted and the density is calculated using image analysis software;

[0090] Blood streaks are divided into regions to ensure statistical accuracy.

[0091] Function: Accurately identify and count traces of blood vessels, providing data support for subsequent health status assessment.

[0092] Detection of the duration of eye closure intervals

[0093] Eye-closed node recognition: Identify the time points of eye-closed nodes through eye tracker data or image analysis;

[0094] Interval duration calculation: Calculate the interval duration for the closed-eye node, and extract the start and end values;

[0095] Time series analysis: Perform time series analysis on the data of eye-closing intervals to assess the dynamic changes in the eye-closing state;

[0096] Function: By dynamically analyzing the duration of eye-closing intervals, the driver's health status can be comprehensively assessed.

[0097] Measurement of fluctuation amplitude when eyes are closed

[0098] Eye movement detection: Detects the number of times the eyes close within a time period using eye tracker data or image transformation.

[0099] Time period calculation: Obtain the initial and end values ​​of the time period, and calculate the time value of the time period;

[0100] Analysis of fluctuation amplitude of eye closure: Dynamic analysis of fluctuation amplitude of eye closure to evaluate the frequency amplitude of the number of times the eyes are closed within a time period;

[0101] Function: To assess the driver's attention and frequency of eye-closing by measuring the amplitude of eye-closing fluctuations.

[0102] Data fusion and preprocessing

[0103] Data extraction: Extract data on the duration of eye-closing intervals, the amplitude of eye-closing fluctuations, and the density of blood vessels from the preprocessed photos;

[0104] Data normalization: Normalize the extracted data to ensure data comparability;

[0105] Data cleaning: Cleaning outliers to ensure the accuracy and reliability of the data;

[0106] Function and purpose: To ensure the quality of the collected data and provide high-quality input data for subsequent model building and health status assessment.

[0107] The communication module is used to upload the data collected by the sensor module and to transmit data, communication and instructions for the entire system.

[0108] The server is used to receive the data transmitted by the communication module, and to analyze and process the data through the model inside the server. It also integrates the analysis structures of the number of blood vessels in the eyeballs, the duration of the eye closure interval, and the fluctuation amplitude of the eye closure to construct a joint analysis of health status change values, comprehensively assess the driver's health status, and generate fatigue assessment information based on the health status change values.

[0109] It should be noted that the server performs feature extraction and analysis on the collected data parameters, mainly analyzing the number of blood vessels in the eyeballs, the duration of the eye-closing interval, and the amplitude of the eye-closing fluctuation. Furthermore, it constructs health status change values ​​based on the number of blood vessels in the eyeballs, the duration of the eye-closing interval, and the amplitude of the eye-closing fluctuation, thereby assessing the driver's health status, facilitating the generation of fatigue assessment information, and improving the driver's driving safety.

[0110] Furthermore, the eye tracker is mainly set to collect the driver's eye-closing points and duration, and through these points, it can calculate the duration of the driver's eye-closing intervals and calculate the amplitude of eye-closing fluctuations based on the frequency of the eye-closing points. This can accurately display the driver's eyelid dynamics and facilitate the assessment of the driver's fatigue level.

[0111] In this embodiment, preferably, the server includes a cloud computing platform, which analyzes and processes data through a model;

[0112] The cloud computing platform also includes an extraction module for extracting features from data to obtain feature sequences, and a model module for analyzing and processing the feature sequences. The model module first analyzes and processes the feature sequences to obtain health status change values, and then the model module analyzes and processes the health status change values ​​to obtain fatigue assessment information.

[0113] It should be noted that the parameters of blood vessels in the eyeball and the parameters of the closed eye node are obtained through the extraction module in the cloud computing platform. The extracted parameters are then combined to form a feature sequence, which is then input into the model module. The model module can calculate and analyze the feature sequence to obtain the health status change value, and then process the health status change value again to assess the driver's fatigue level.

[0114] In this embodiment, preferably, the steps for the cloud computing platform to analyze and process the transmitted data information are as follows:

[0115] The server receives and stores data: it receives and stores the data transmitted by the communication module, and stores the received data to the cloud computing platform to form a complete dataset of driver health status.

[0116] Data analysis and model building: Technical features were extracted from the driver health status dataset, and model modules were built for the number of blood vessels in the eyes, the duration of the eye closure interval, and the amplitude of fluctuations when the eyes are closed.

[0117] Health status data fusion and analysis: The driver health status dataset is fused and analyzed to obtain a joint analysis of the number of blood vessels in the eyeballs, the duration of the eye closure interval, and the fluctuation amplitude of the eye closure, and to obtain the health status change value H;

[0118] The model module calculates the change in health status: analyzes the range of numerical parameters of the change in health status H, and assesses the driver's health status.

[0119] Fatigue assessment generation: The model module is used to calculate the health status change value H. Based on the change of the health status change value H, the driver's fatigue level is assessed and fatigue assessment information is output.

[0120] It should be noted that by storing the received data on a cloud computing platform, it is easier to store the data, form an effective dataset, facilitate accurate data querying and processing, and improve data tracking and analysis.

[0121] Furthermore, the driver's health status dataset is analyzed and processed through the model module to facilitate the calculation of the parameter range of the output health status change value H, and the driver's fatigue is assessed based on the health status change value H.

[0122] Furthermore, the health status change value H is processed by combining the number of blood vessels in the eyeballs, the duration of the eye-closing interval, and the amplitude of the eye-closing fluctuation, which facilitates effective calculation and analysis. By analyzing and calculating the blood vessels in the eyeballs and the eye-closing state, a comprehensive evaluation can be achieved.

[0123] In this embodiment, preferably, the feature extraction steps for the driver health status dataset are as follows:

[0124] First, the preprocessed data of the eyeball image is planarized to maintain the planarity of the preprocessed eyeball image;

[0125] The image was cropped using image editing software to remove the black part of the eye; the cropped image only retains the white part of the eye, clearly showing the bloodshot marks.

[0126] Color channel-based blood vessel detection identifies blood vessels in the sclera (white of the eye); the results include the number and region of blood vessels; and the density of blood vessels within each region is calculated.

[0127] It should be noted that by flattening the eyeball image, the eyeball image can be made easier to analyze and process, and the black part of the eyeball can be cropped to facilitate color channel analysis of the white part of the eyeball, blood vessels can be cleaned and blood vessel data can be statistically analyzed, and the flattened eyeball image can be divided into regions to facilitate the calculation of blood vessel density.

[0128] ;

[0129] in, Expressed as blood vessel density, This represents the number of blood vessels within the area. Represented as the area of ​​the divided region. This represents the total number of regions, and Represented as a sequence of regions, It represents the average density of blood vessels within the divided area.

[0130] In this embodiment, preferably, the analysis of the duration of the eye-closing interval is as follows:

[0131] Preprocessed data on the duration of eye closure intervals monitored by eye-tracking devices are subjected to time-series processing;

[0132] The closed-eye nodes are linked to the time series, and features are extracted from the closed-eye nodes in the time series.

[0133] The technical feature extraction for the data of eye-closing interval duration includes the start and end values ​​of the interval duration in the time series, and then calculates the eye-closing interval duration and the fluctuation amplitude of eye-closing.

[0134] It should be noted that by performing time series processing on the nodes in the closed-eye state, it is easier to extract features from the closed-eye nodes, and to calculate the duration of the closed-eye interval based on the end and start values ​​of the closed-eye time, as well as to calculate the fluctuation amplitude of the closed-eye state through the time series.

[0135] The calculation of the interval between eye closures is as follows:

[0136] ;

[0137] ;

[0138] in, Represented as a data sequence of eye-closed duration, and This represents the number of closed-eye nodes. This represents the minimum duration of the eye-closing interval. This represents the duration of time the eyes are closed. This represents the starting value of the time the eyes are closed. This represents the end value of the time the eyes are closed;

[0139] The amplitude of fluctuation when eyes are closed is calculated as follows:

[0140] ;

[0141] in, This is expressed as the amplitude of fluctuation when eyes are closed. This represents the number of times the eyes are closed within a given time period, and Indicates the distinguishing character, This represents the end time value of a time period. This represents the beginning time value of a time period.

[0142] In this embodiment, preferably, the formula for calculating the health status change value H is as follows:

[0143] ;

[0144] in, Represented as a change in health status. This represents the duration of eye closure, and This is represented as a sequence of positions in the dataset of eye-closing intervals. This represents the minimum duration of the eye-closing interval. This is expressed as the amplitude of fluctuation when eyes are closed, and This is represented as the position sequence in the closed-eye fluctuation amplitude dataset. This represents the maximum amplitude of the fluctuation when the eyes are closed. Expressed as blood vessel density, , , and All are represented as model parameters;

[0145] It should be noted that the model parameters were determined through a large amount of experimental data. , , and The duration of eye closure was analyzed using a multivariate nonlinear regression algorithm. Fluctuation amplitude when eyes are closed and blood vessel density Modeling; and A nonlinear transformation is applied to the duration of the eye-closing interval to enhance the flexibility of the formula. The normalization process for the fluctuation amplitude when eyes are closed avoids the influence of dimensional differences, resulting in a better output of the health status change value. The range of its value is in the range (0,1); when When the value approaches 0, it indicates a stable health status; when... When the value approaches 1, it indicates an abnormal health condition or fatigue.

[0146] This embodiment demonstrates the innovation and advantages of an intelligent logistics monitoring and early warning system based on a cloud computing platform in practical applications. Through multi-dimensional analysis of the duration of eye closure intervals, the amplitude of eye closure fluctuations, and the density of blood vessels, the system can accurately assess the driver's health status, generate targeted early warning information, help drivers detect and respond to fatigue in a timely manner, and significantly improve the safety and efficiency of logistics transportation.

[0147] The terminal module is used to integrate data from multiple modules, output comprehensive and accurate early warning information based on fatigue assessment information, and construct early warning levels.

[0148] Furthermore, the health status change value H generates fatigue assessment information as follows:

[0149] ;

[0150] in, This represents a classification of fatigue levels. Represented as a change in health status, ranging from [0, 1]; when When the value approaches 0, it indicates a stable health status, corresponding to a low fatigue level; when... When the value approaches 1, it indicates an abnormal health condition or extreme fatigue, corresponding to a severe fatigue level.

[0151] Through changes in health status values The multi-level assessment comprehensively evaluates driver fatigue, providing a basis for generating early warning information;

[0152] It should be noted that the piecewise function design ensures the accuracy and operability of the classification, with each level corresponding to a change in health status. The range reflects the characteristics of different health states, comprehensively assesses driver fatigue, provides a basis for generating early warning information, and when When the value approaches 0, it indicates a stable health status, corresponding to a low fatigue level; when... When the value approaches 1, it indicates an abnormal health condition or extreme fatigue, corresponding to a severe fatigue level.

[0153] Setting up the test subjects:

[0154] Driver A: Long-distance logistics driver, working hours 4 hours, in good health.

[0155] Driver B: Long-distance logistics driver, working 6-hour shifts, experiencing mild fatigue symptoms.

[0156] Driver C: Long-distance logistics driver, working 8-hour shifts, experiencing moderate fatigue symptoms.

[0157] Driver D: Long-distance logistics driver, working 10-hour shifts, experiencing severe fatigue symptoms.

[0158] The above calculation formula was tested, and the experimental results are shown in Table 1.

[0159] test subjects Duration of eye-closing interval Blood density Fluctuation amplitude when eyes are closed Health status change value Fatigue level Driver A 2.8 0.12 0.8 0.15 Low Driver B 4.1 0.18 1.2 0.65 moderate Driver C 6.3 0.25 1.5 0.75 serious Driver D 8.2 0.35 1.8 0.85 serious

[0160] Icon Analysis

[0161] Duration of eye-closing interval

[0162] Data Comparison:

[0163] Driver A's eye-closing interval was 2.8 seconds, which is relatively short and reflects good health.

[0164] Driver B's eye-closing interval lasted 4.1 seconds, which is slightly long and may reflect mild fatigue.

[0165] Driver C's eye-closing interval lasted 6.3 seconds, which is relatively long and indicates moderate fatigue.

[0166] Driver D's eye-closing interval lasted 8.2 seconds, which is significantly long and reflects severe fatigue.

[0167] Blood density

[0168] Data Comparison:

[0169] Driver A's blood density was 0.12, which is low and indicates good blood circulation;

[0170] Driver B's blood vessel density is 0.18, which is moderate and may reflect mild fatigue;

[0171] Driver C's blood vessel density was 0.25, which is relatively high, indicating moderate fatigue;

[0172] Driver D's blood vessel density was 0.35, which is relatively high and indicates severe fatigue.

[0173] Fluctuation amplitude when eyes are closed

[0174] Data Comparison:

[0175] Driver A's eye-closing fluctuation range was 0.8 degrees, which is relatively small and reflects that the eye muscles are in good condition;

[0176] Driver B's eye-closing fluctuation range was 1.2 divisions, which is slightly large and may reflect mild fatigue;

[0177] Driver C's eye-closing fluctuation range was 1.5 degrees, which is relatively large and reflects moderate fatigue;

[0178] Driver D's eye-closing fluctuation range was 1.8 degrees, which is relatively large and reflects severe fatigue;

[0179] Health status change value

[0180] Data Comparison:

[0181] Driver A's =0.15, low, reflecting a stable health status;

[0182] Driver B's =0.65, generally speaking, reflects mild fatigue;

[0183] Driver C's =0.75, moderate, reflecting moderate fatigue;

[0184] Driver D's =0.85, severe, reflecting severe fatigue.

[0185] Correlation between fatigue level and early warning information:

[0186] when When the value approaches 0, it indicates a stable health condition, corresponding to a low fatigue level, suggesting the driver rest for 5 minutes.

[0187] when When the value approaches 1, it indicates an abnormal health condition or extreme fatigue, corresponding to a severe fatigue level, and it is recommended that the driver stop driving immediately.

[0188] Quantitative analysis of technical effects:

[0189] Early warning accuracy: based on changes in health status values According to the analysis, the system can accurately identify the driver's fatigue state, with a warning accuracy rate of over 90%.

[0190] Response speed: The system collects data and performs calculations Afterwards, the average response time is within 5 seconds, ensuring the timeliness of early warning information;

[0191] Significant technological benefits: Through multi-dimensional analysis of health status data, the system can significantly improve the accuracy and response speed of early warnings, reduce traffic accidents caused by driver fatigue, and improve the overall safety and efficiency of logistics transportation.

[0192] The notification module is used by the terminal module to communicate, contact, and remind logistics drivers based on fatigue assessment information, and to force logistics drivers to take rest.

[0193] Standard communication protocols (such as TCP / IP, HTTP, MQTT) are used to enable data interaction between the terminal module and the vehicle-mounted equipment; the driver is forced to rest or stop working through API interfaces or command line commands; the status information of the terminal module is synchronized regularly to ensure the timeliness and accuracy of the notification content;

[0194] Notification content type:

[0195] Common reminders: such as "It is recommended to rest for 5 minutes".

[0196] Warning message: such as "Fatigue level is moderate, it is recommended to rest immediately".

[0197] Urgent Notice: Such as "Abnormal health status or extreme fatigue, corresponding to severe fatigue level".

[0198] Personalized customization: Based on the driver's historical data and work habits, the notification content is customized to improve the reminder effect.

[0199] Technical means to implement forced rest:

[0200] Terminal device control: Force entry into rest mode via the terminal module's API interface.

[0201] Sound and vibration alerts: Use sound alerts and phone vibration to ensure drivers notice notifications.

[0202] Combining multiple methods: Combining various reminder methods such as sound, vibration, text message, and push notifications to enhance the effectiveness of forced rest.

[0203] It should be noted that all calculation formulas in this application employ regression analysis, including but not limited to machine learning algorithms, to deeply analyze the collected parameters and identify their natural trends and interrelationships. Specialized software, such as Python's Scikit-learn library or the R language, is used to automatically generate mathematical models that match the data. Then, cross-validation and other methods are used to objectively evaluate the model performance, and continuous feedback and optimization are combined to ensure that the created formulas truly reflect the inherent laws of the data, thereby guaranteeing their effectiveness and accuracy. In all calculation formulas in this application, the parameters in each formula undergo dimensionless processing within a consistent range to ensure that different physical quantities are compared on the same scale; dimensionless processing techniques include, but are not limited to, Min-Max Normalization and Z-Score standardization.

[0204] The technical solution of this invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of this invention.

[0205] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0206] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An intelligent logistics monitoring and early warning system based on a cloud computing platform, characterized in that, The sensor module, the data acquisition module, the data preprocessing module, the communication module, the server, the terminal module and the notification module are included. The sensor module is used for monitoring the logistics driver, and the collected data includes taking eye photos, eye closure interval length and eye closure fluctuation amplitude. The communication module is used for uploading the data collected by the sensor module, and realizing the transmission of data, communication and instructions of the whole system. The server is used for receiving the data transmitted by the communication module, and analyzing and processing the data through the model inside the server. The eye photos are processed to calculate the blood filament density: the preprocessed data of the eye photos are planarized to keep the planarity of the preprocessed eye photos; the planar eye photos are cropped using an image editing software to remove the black eye area; only the white eye part is reserved after cropping, and the blood filament traces are clearly displayed; the blood filament traces in the white eye part are identified based on the color channel blood filament detection; and the identification result includes the number and area of the blood filaments. The density of the blood filaments in the area is calculated; and the analysis structures of the number of eye blood filaments, eye closure interval length and eye closure fluctuation amplitude are fused to construct a joint analysis health state change value, and the health state of the driver is comprehensively evaluated, and the fatigue degree evaluation information is generated according to the health state change value. The calculation formula of the health state change value is as follows: ； wherein, is expressed as a change in health state value, is expressed as a length of closed eyes, is expressed as a sequence of positions in a dataset of length of closed eyes breaks, is expressed as a minimum of length of closed eyes breaks, is expressed as a fluctuation amplitude of closed eyes, is expressed as a sequence of positions in a dataset of fluctuation amplitude of closed eyes, is expressed as a maximum of fluctuation amplitude of closed eyes, is expressed as a density of blood vessels, , , and are each expressed as a model parameter; The terminal module is used for integrating the data of multiple modules, outputting comprehensive and accurate early warning information according to the fatigue degree evaluation information, and constructing an early warning level. 2.The cloud computing platform-based intelligent logistics monitoring and early warning system according to claim 1, characterized in that: The sensor module includes an eye imager of an eye imaging device and an eye tracker of an eye movement monitoring device. The sensor module is arranged at the front of the logistics vehicle and at the upper end of the driving position to realize taking eye photos and monitoring eye closure interval length of the logistics driver. The eye imager is used for taking eye photos of the driver, and the eye movement monitoring device is used for monitoring the eye closure interval length and the eye closure fluctuation amplitude of the driver. 3.The cloud computing platform-based intelligent logistics monitoring and early warning system according to claim 2, characterized in that: The data acquisition module is integrated in the inside of the sensor module, and is used for controlling the eye imager and the eye tracker to take eye photos and monitor the eye closure interval length and the eye closure fluctuation amplitude. The data preprocessing module is also integrated in the inside of the sensor module, and is used for preprocessing the eye photos and the monitored eye closure interval length collected by the data acquisition module to obtain preprocessed data. 4.The cloud computing platform-based intelligent logistics monitoring and early warning system according to claim 3, characterized in that: The server includes a cloud computing platform, which analyzes and processes data through a model. The cloud computing platform further includes an extraction module for feature extraction to obtain a feature sequence, and a model module for analyzing and processing the feature sequence, which first analyzes and processes the feature sequence to obtain a health state change value, and then analyzes and processes the health state change value to obtain fatigue degree evaluation information. 5.The cloud computing platform-based intelligent logistics monitoring and early warning system according to claim 4, characterized in that: The steps of analyzing and processing the transmitted data information by the cloud computing platform are as follows: The server receives and stores the data transmitted by the communication module, stores the received data to the cloud computing platform, and forms a complete driver health state data set; Data analysis and model construction: technical feature extraction is performed on the driver health state data set, and model modules of the number of blood vessels in the eyeball, the length of the closed-eye interval, and the closed-eye fluctuation amplitude are constructed; Health status data fusion and analysis: the driver health state data set is fused and analyzed to obtain the joint analysis of the number of blood vessels in the eyeball, the length of the closed-eye interval, and the closed-eye fluctuation amplitude, and obtain the health state change value H; Calculate the health state change value through the model module: analyze the numerical parameter range of the health state change value H to evaluate the health state of the driver; Fatigue assessment generation: calculate the health state change value H using the model module, evaluate the fatigue of the driver according to the change of the health state change value H, and output the fatigue assessment information. 6.The cloud computing platform-based intelligent logistics monitoring and early warning system according to claim 5, characterized in that: The analysis of the closed-eye interval length is as follows: The pre-processed data of the closed-eye interval length of the driver monitored by the eye movement monitoring device is time sequenced; The closed-eye node is associated with the time sequence, and the features of the closed-eye node under the time sequence are extracted; The technical feature extraction of the closed-eye interval length data feature extraction includes the start value and end value of the time sequence of the interval length, and then the closed-eye interval length and the closed-eye fluctuation amplitude are calculated. 7.The cloud computing platform-based intelligent logistics monitoring and early warning system according to claim 6, characterized in that: The health state change value H generates fatigue assessment information as follows: ； in, This represents a classification of fatigue levels. Represented as a change in health status, ranging from [0, 1]; when When the value approaches 0, it indicates a stable health status, corresponding to a low fatigue level; when... When the value approaches 1, it indicates an abnormal health condition or extreme fatigue, corresponding to a severe fatigue level. By the health state change value The driver's fatigue degree is comprehensively evaluated by the multi-stage evaluation of the health state change value, so as to provide a basis for the generation of early warning information.

Images