Personnel riding monitoring management method and system

Abstract

The present application relates to the field of traffic detection, and in particular to a method and system for monitoring and managing riding of people, which includes: image acquisition: acquiring monitoring images; behavior detection: judging whether the monitoring images meet the detection conditions, and executing the information extraction step according to the judgment result; information extraction: determining the identification information based on the monitoring images; age estimation: determining the target age information based on the identification information; alarm detection: judging whether the target age information meets the age condition, and executing the riding warning step according to the judgment result. When the monitoring image detects that someone is riding a vehicle, the skeleton information will be extracted through the monitoring image. The skeleton information can reflect the figure of the rider, and the age of the rider can be estimated by the figure of the rider. When the rider is young, a riding warning will be triggered. Using the body proportions of the rider, the system can estimate the age of the rider more accurately and warn young riders in time.

Description

Technical Field

[0001] The present application relates to the field of traffic detection, and in particular to a method and system for monitoring and managing riding of personnel. Background Art

[0002] As people's awareness of green travel increases, there are vehicles such as shared bicycles, electric bicycles, and shared bicycles on the market. These vehicles are light, convenient, cheap, and easy to park. The amount and scope of investment in urban transportation are also increasing. In order to improve traffic safety measures during transportation, vehicles need to limit the minimum age of riders, such as prohibiting young people under 12 from riding shared bicycles. However, for shared vehicles such as shared bicycles and shared bicycles, their unlocking method is generally achieved through APP, which is difficult to effectively restrict young people from unlocking, and has limitations.

[0003] In the prior art, for example, a Chinese invention application with application publication number CN108454740A discloses a detection and control device for a shared bicycle, which includes a pressure sensor and a 51 single-chip microcomputer. The pressure sensor is used to detect the weight data of a person riding a shared bicycle. The built-in program of the 51 single-chip microcomputer can analyze the weight data to determine whether the rider meets the weight requirement, and the program will only be executed when the rider's weight reaches or exceeds the average weight of a 12-year-old.

[0004] Regarding the above technical solution, the inventor believes that when young people ride shared bicycles with heavy objects on their backs, the weight data measured by the pressure sensor is significantly different from the actual weight data of the young people. It is difficult for the system to accurately estimate the age of the young people and it is difficult to warn the young people in time. Summary of the invention

[0005] The first purpose of this application is to provide a method for monitoring and managing riding of personnel, which has the characteristic of improving detection accuracy.

[0006] The above-mentioned invention objective of the present application is achieved through the following technical solutions:

[0007] Personnel riding monitoring and management methods include:

[0008] Image acquisition: obtain surveillance images;

[0009] Behavior detection: judging whether the monitoring image meets the detection condition, and executing the information extraction step according to the judgment result; wherein the detection condition is that the monitoring image records a vehicle and a rider riding on the vehicle;

[0010] Information extraction: determining identification information based on the surveillance image; wherein the identification information includes skeleton information, and the skeleton information is used to reflect the body shape of the rider;

[0011] Age estimation: Determine the target age information based on identification information;

[0012] Alarm detection: Determine whether the target age information meets the age requirement, and execute the riding warning steps according to the judgment result.

[0013] By adopting the above technical solution, when the monitoring image detects a person riding a vehicle, the skeleton information will be extracted from the monitoring image. The skeleton information can reflect the figure of the rider, and the age of the rider can be estimated by the figure of the rider. When the rider is young, a riding warning will be triggered. Using the body proportions of the rider, the system can estimate the age of the rider more accurately and warn young riders in time.

[0014] Optionally, the specific method of the behavior detection step includes:

[0015] Based on the monitoring image, determine a human body area and a vehicle body area; wherein the human body area is used to indicate the rider, and the vehicle body area is used to indicate the vehicle;

[0016] Based on the human body area and the vehicle body area, determine the overlapping area;

[0017] An area comparison is performed based on the overlapping area. If the area of ​​the overlapping area is greater than the area threshold, the monitoring image meets the detection condition and the information extraction step is executed.

[0018] By adopting the above technical solution, the overlapping area is obtained based on the human body area and the vehicle body area, and a comparison and judgment is made based on the overlapping area and the area threshold to detect whether the person is in a riding state. When the person is not in a riding state, no subsequent calculation is performed, thereby reducing the probability of analysis errors.

[0019] Optionally, the specific method for determining identification information based on the surveillance image includes:

[0020] Determine a target area image based on the human body area, the vehicle body area and the monitoring image; wherein the target area image can reflect the rider and the vehicle;

[0021] Perform perspective transformation based on the target area image to determine the optimized image;

[0022] Based on the optimized image, skeleton information is determined.

[0023] By adopting the above technical solution, perspective transformation can alleviate the inclination problem when the image is taken, and improve the accuracy of subsequent data analysis.

[0024] Optionally, in the age estimation step, include:

[0025] The skeleton information is substituted into the age estimation model to determine the target age information; wherein the skeleton information includes arm length information reflecting the arm length of the rider, leg length information reflecting the leg length of the rider, and spine information reflecting the spine length of the rider.

[0026] By adopting the above technical solution, the skeleton information can reflect the rider's arm length, leg length and spine length, more comprehensively reflect the rider's body skeleton, and improve the accuracy of age estimation in subsequent analysis.

[0027] Optionally, the method for generating the age estimation model includes:

[0028] Determine fitting information; wherein the fitting information includes age fitting information corresponding to people of different age groups, arm length fitting information corresponding to people of different age groups, leg length fitting information corresponding to people of different age groups, and spine fitting information corresponding to people of different age groups; the age fitting information is used to reflect the age group of the person;

[0029] The fitting information was substituted into the multivariate function model for orthogonal polynomial fitting to generate an age estimation model.

[0030] By adopting the above technical solution and performing curve fitting on the age estimation model through orthogonal polynomials, the influence of errors on the independent variables and dependent variables in the fitting process can be reduced, making the age estimation model more reasonable.

[0031] Optionally, the specific method of the information extraction step includes:

[0032] Determine facial information based on surveillance images;

[0033] Among the specific methods of age estimation are:

[0034] Based on the identification information, determining first age information;

[0035] Determining second age information based on the facial information;

[0036] Based on the first age information and the second age information, target age information is determined.

[0037] By adopting the above technical solution, facial information can reflect the rider's face. The rider's age can be estimated by analyzing the rider's facial data. The age estimated by combining the skeleton information and the age estimated by the facial information can be combined with the rider's figure and face to estimate the age, thereby improving the accuracy of age estimation.

[0038] Optionally, the specific method of the riding warning step includes:

[0039] Send facial information to the operator;

[0040] The operator compares the facial information in the database to determine the identity information;

[0041] The operator sends a warning message to the user terminal corresponding to the identity information.

[0042] By adopting the above technical solution, the operator can send a warning message to the user end, timely lock the user of the rider and the means of transportation being used, and warn the rider.

[0043] The second purpose of this application is to provide a personnel riding monitoring and management system with the characteristic of improving detection accuracy.

[0044] The second object of the invention is achieved by the following technical solutions:

[0045] An image acquisition module, used to acquire monitoring images;

[0046] A behavior detection module, used to determine whether the monitoring image meets the detection condition, and execute the information extraction step according to the determination result; wherein the detection condition is that the monitoring image records a vehicle and a rider riding on the vehicle;

[0047] An information extraction module, used to determine identification information based on the monitoring image; wherein the skeleton information is used to reflect the body shape of the rider;

[0048] An age estimation module, used to determine target age information based on the identification information;

[0049] The alarm detection module is used to determine whether the target age information meets the age requirement and execute the riding warning step according to the determination result.

[0050] The third objective of this application is to provide an intelligent terminal with the characteristic of improving detection accuracy.

[0051] The third object of the invention is achieved by the following technical solutions:

[0052] An intelligent terminal includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the above-mentioned personnel riding monitoring management method.

[0053] The fourth objective of this application is to provide a computer storage medium that can store corresponding programs and has the characteristic of improving detection accuracy.

[0054] The fourth object of the invention of this application is achieved through the following technical solutions:

[0055] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any of the above-mentioned personnel riding monitoring management methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a flow chart of the personnel riding monitoring and management method of the first embodiment of the present application.

[0057] Figure 2 It is a schematic diagram of the image acquisition module collecting monitoring images.

[0058] Figure 3 It is a sub-process diagram of the personnel riding monitoring management method of Example 1 of the present application.

[0059] Figure 4 It is a flow chart of the personnel riding monitoring and management method of the second embodiment of the present application.

[0060] Figure 5 It is a module diagram of the personnel riding monitoring and management system of the present application.

[0061] In the figure, 1. Image acquisition module; 2. Behavior detection module; 3. Information extraction module; 4. Age estimation module; 5. Alarm detection module; 6. Riding warning module. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0063] In addition, the numbers of the steps in this embodiment are only for convenience of explanation and do not limit the execution order of the steps. In practical applications, the execution order of the steps can be adjusted as needed, or performed simultaneously. These adjustments or replacements are all within the scope of protection of the present invention.

[0064] The following is attached to the instruction manual Figure 1-Figure 5 The embodiments of the present application are described in further detail.

[0065] Embodiment 1:

[0066] An embodiment of the present application provides a method for monitoring and managing riding of personnel, and the main process of the method is described as follows.

[0067] Reference Figure 1 and Figure 2 , S1. Image acquisition: obtain monitoring images.

[0068] The monitoring image is obtained by photographing the traffic section by the image acquisition module 1. In this embodiment, the image acquisition module 1 is a monitoring camera arranged at a traffic post.

[0069] Specifically, the monitoring image is restored by Wiener filtering before entering step S2 to reduce noise interference in the image and improve the accuracy of image data extraction in subsequent steps.

[0070] S2, behavior detection: determine whether the monitoring image meets the detection conditions, and execute S3 according to the judgment result.

[0071] The detection condition is that the monitoring image records both the means of transport and the riders riding the means of transport. The monitoring image is obtained based on the shooting of public traffic sections. There may be people who are not riding the means of transport in the monitoring image, or there may be means of transport without people riding. Subsequent data analysis can only be performed when the riders are riding public transportation. In this embodiment, the means of transport is a shared bicycle.

[0072] Reference Figure 1 and Figure 3 , in step S2, including:

[0073] S21. Determine a human body area and a vehicle body area based on the monitoring image.

[0074] In the monitoring image, the rider and the vehicle are identified respectively, and the identified local area corresponding to the rider is cut out to obtain the human body area, and the identified local area corresponding to the vehicle is cut out to obtain the vehicle body area. Therefore, the human body area can indicate the rider, and the vehicle body area can indicate the vehicle.

[0075] S22: Determine an overlapping area based on the human body area and the vehicle body area.

[0076] The human body regions and the vehicle body regions correspond one to one, and the overlap region refers to the overlapped portion between one of the human body regions and the corresponding vehicle body region.

[0077] S23, determine whether the area of ​​the overlapping region is greater than the area threshold, if so, execute S3; if not, end.

[0078] Among them, by adding the human body area and the vehicle body area to the preset reference coordinate system respectively, the coordinates of each corner point of the human body area, the vehicle body area and the overlapping area can be obtained respectively, and then the areas of the human body area, the vehicle body area and the overlapping area can be calculated respectively.

[0079] When the area of ​​the overlapped region is greater than the area threshold, it means that the overlap between the rider and the vehicle is high, and the rider is more likely to use the vehicle. Therefore, it can be determined that the rider corresponding to the human body region is riding the vehicle corresponding to the vehicle body region, and the detection condition is met, and subsequent age estimation can be performed. Otherwise, it can be determined that the rider corresponding to the human body region is not riding the vehicle corresponding to the vehicle body region, and the detection condition is not met, and subsequent age estimation is not required.

[0080] Specifically, the area threshold is associated with the overlap coefficient and the total area, wherein the total area is the sum of the areas of the human body area and the vehicle body area, and the area threshold is the product of the overlap coefficient and the total area. In this embodiment, the overlap coefficient is preferably 0.7.

[0081] S3. Information extraction: Determine identification information based on surveillance images.

[0082] Among them, identification information refers to the information extracted from the surveillance image that can reflect the human body.

[0083] Reference Figure 1 and Figure 3 , in step S3, including:

[0084] S31. Determine skeleton information based on the monitoring image.

[0085] The skeleton information is used to reflect the body shape of the rider.

[0086] In step S31, it includes:

[0087] S311, determining a target area image based on a human body area, a vehicle body area and a monitoring image.

[0088] Among them, in the monitoring image, the rider corresponding to the human body area and the traffic vehicle corresponding to the vehicle body area are combined and cut out to obtain a target area image, in which the rider and the vehicle the rider is riding are recorded.

[0089] S312: Perform image correction based on the target area image to determine an optimized image.

[0090] Among them, image correction includes perspective transformation; perspective transformation can perform perspective correction on the current area image to obtain an optimized image.

[0091] Reference Figure 2 When the image acquisition module 1 is collecting monitoring images, it is difficult for the image acquisition module 1 to face the vehicle or the rider. In this embodiment, the image acquisition module 1 uses a monitoring camera as an example. Due to the installation height and elevation angle of the monitoring camera, there will be pixel offset in the monitoring image, resulting in a large difference between the ratio of people to vehicles in the current area image and the actual ratio of people to vehicles. Therefore, it is necessary to perform perspective transformation on the current area image.

[0092] The specific method of perspective transformation is:

[0093] Substitute the pixel coordinates of each pixel point of the current region image into formula (1) and formula (2) to obtain the pixel coordinates of the pixel point after perspective transformation.

[0094] (1)

[0095] (2)

[0096] Among them, the pixel coordinates of the pixel point in the current area image are (u, v, w), and w=1;

[0097] The pixel coordinates of the pixel after perspective transformation are (x', y', w'), and w'=1;

[0098] a 11 、a 12 、a 13 , …, a 33 Conversion parameters set for the system.

[0099] Specifically, the perspective transformation process also includes:

[0100] Identify the front of the vehicle in the current area image, and determine whether the direction of the front of the vehicle in the current area image is consistent with the preset standard direction. If so, execute the next step; if not, correct and identify the direction of the front of the vehicle in the current area image to make the direction of the front of the vehicle in the optimized image consistent with the preset standard direction.

[0101] The vehicle head refers to the head of the vehicle in the current area image, and the vehicle head direction refers to the direction of the vehicle head. Taking the standard direction as from left to right as an example, if the vehicle head direction in the current area image is from right to left, then during the perspective transformation process, the vehicle head direction in the current area image also needs to be changed to from left to right, so that the vehicle head direction in the optimized image is consistent with the preset standard direction.

[0102] Since the direction of the vehicle head in each optimized image is consistent with the preset standard direction, the position of the vehicle head in the optimized image can be roughly fixed, thereby enhancing the accuracy of subsequent judgments.

[0103] Determine the standard size and make the optimized image size consistent with the standard size.

[0104] The standard size is a standard value preset by the system, and the size of each optimized image needs to be consistent with the standard size, thereby standardizing the sizes of all optimized images.

[0105] If the size of the optimized image is too small, in the subsequent steps, if the optimized image is directly imported into the analysis model, it may cause loss of image information. Using standard size specifications to optimize the image size can reduce the loss caused by the small size of the current regional image after clipping.

[0106] Reference Figure 1 and Figure 3 , S313, determine skeleton information based on the optimized image.

[0107] By performing human skeleton recognition on the optimized image, skeleton information corresponding to the rider can be obtained, and the skeleton information can reflect the human skeleton of the rider. In this embodiment, the skeleton information includes arm length information, leg length information and spine information, wherein the arm length information can reflect the arm length, the leg length information can reflect the leg length, and the reference information can reflect the spine length.

[0108] S4. Age estimation: Determine the target age information based on the identification information.

[0109] The target age information is used to indicate the estimated result of the age of the rider. The identification information includes skeleton information. By analyzing the skeleton information, the arm length, leg length and reference length of the rider can be analyzed to estimate the age of the rider.

[0110] Reference Figure 3 , in step S4, including:

[0111] S41. Determine first age information based on the skeleton information.

[0112] Among them, by substituting the skeleton information into the age estimation model, the age of the current rider can be estimated.

[0113] Specifically, the age estimation model is a curve model, which is obtained by fitting the fitting information through a multiple linear regression method.

[0114] The generation method of the age estimation model includes:

[0115] 1) Determine the fitting information.

[0116] The fitting information includes age fitting information corresponding to people of different age groups, arm length fitting information corresponding to people of different age groups, leg length fitting information corresponding to people of different age groups, and spine fitting information corresponding to people of different age groups; the age fitting information is used to reflect the age group of the person. The fitting information is equivalent to collecting the arm length, leg length, and spine length of people of different age groups.

[0117] 2) Substitute the fitting information into the multivariate function model for orthogonal polynomial fitting to generate an age estimation model.

[0118] Among them, the collected arm length, leg length and spine length of people of different age groups are used as input, and the corresponding age group of the people is used as output. Orthogonal polynomial fitting is performed to fit a multivariate function, namely the age estimation model.

[0119] S42: Determine target age information based on the first age information.

[0120] Among them, the age corresponding to the target age information is the age estimated in the first age information, completing the estimation of the rider's age.

[0121] S5, alarm detection: determine whether the target age information meets the age limit condition, if yes, end; if not, execute S6.

[0122] Among them, the age condition refers to whether the age corresponding to the target age information is greater than or equal to the age threshold, and the age threshold is the lower age limit preset by the system. Taking the lower age limit as 12 years old as an example, the age threshold is 12; when the age value corresponding to the target age information is greater than or equal to 12, the age of the rider meets the age condition and the rider can use the vehicle normally; when the age value corresponding to the target age information is less than 12, the age of the rider does not meet the age condition and the rider needs to be warned, triggering the riding warning step.

[0123] S6. Riding warning.

[0124] Among them, warnings will be given to riders who do not meet the age requirements, reminding them to stop riding in time.

[0125] The implementation principle of the first embodiment of the present application is as follows: when the monitoring image detects that a person is riding a vehicle, the skeleton information will be extracted through the monitoring image, and the skeleton information can reflect the arm length, leg length and spine length of the rider. In combination with the arm length, leg length and spine length, the rider's age can be comprehensively estimated using the rider's stature, thereby improving the accuracy of age estimation in subsequent analysis. When the rider is young, a riding warning will be triggered. Using the body proportions of the rider, the system can estimate the rider's age more accurately and warn young riders in a timely manner.

[0126] On the other hand, the image acquisition module 1 uses the monitoring camera preset at the traffic post. In the process of acquiring image information, there is no need to add additional external hardware equipment to the vehicle body, which is more convenient.

[0127] Embodiment 2:

[0128] The embodiment of the present application provides a method for monitoring and managing riding of personnel, which is different from the first embodiment in that:

[0129] Reference Figure 4 , in step S3, further comprising:

[0130] S32. Determine facial information based on the surveillance image.

[0131] The identification information also includes skeleton information and face information. The face information refers to an image extracted from the monitoring image that can reflect the human face. By performing face recognition on the optimized image, the face information corresponding to the rider can be obtained.

[0132] In step S4, it also includes:

[0133] S42: Determine second age information based on the facial information.

[0134] Among them, by inputting the facial image in the facial information into a preset neural network, the age of the rider can be estimated and the second age information can be determined.

[0135] In step S43, target age information is determined based on the first age information and the second age information.

[0136] The first age information is the estimated age obtained by analyzing the human skeleton, and the second age information is the estimated age obtained by analyzing the human face. By combining the first age information and the second age information, the target age information can be obtained.

[0137] In this embodiment, the age value corresponding to the target age information is an average value between the age value of the first age information and the age value of the second age information.

[0138] In step S6, it includes:

[0139] S61. Send the facial information to the operator.

[0140] Among them, in this embodiment, taking the means of transportation as a shared bicycle as an example, the operator refers to the operator of the means of transportation.

[0141] S62: The operator compares the facial information in the database to determine the identity information.

[0142] Among them, the facial image in the captured facial information is uploaded to the operator's database for comparison, so as to determine the identity of the user corresponding to the facial information, thereby determining the identity information.

[0143] S63: The operator sends a warning message to the user terminal corresponding to the identity information.

[0144] The implementation principle of the second embodiment is: facial information can reflect the face of the rider. The rider's age can be estimated by analyzing the rider's facial data. The age estimated by combining the age estimated by the skeleton information and the age estimated by the facial information can be combined with the rider's body shape and face to estimate the age, thereby improving the accuracy of age estimation.

[0145] Embodiment three:

[0146] In one embodiment, a personnel riding monitoring management system is provided, which corresponds one-to-one with the personnel riding monitoring management method in the above-mentioned embodiment 1. Figure 5 ,The personnel riding monitoring and management system includes an image acquisition module 1, a behavior detection module 2, an information extraction module 3, an age estimation module 4, an alarm detection module 5 and a riding warning module 6. The detailed description of each functional module is as follows:

[0147] The image acquisition module 1 is used to acquire monitoring images and send monitoring image information to the behavior detection module 2.

[0148] The behavior detection module 2 is used to determine whether the monitoring image meets the detection conditions, execute the information extraction step according to the judgment result, and send the image detection information to the information extraction module 3; wherein the detection condition is that the monitoring image records the transportation vehicle and the rider riding on the transportation vehicle.

[0149] The information extraction module 3 is used to determine the identification information based on the monitoring image and send the skeleton information to the age estimation module 4; wherein the identification information includes the skeleton information, and the skeleton information is used to reflect the body shape of the rider.

[0150] The age estimation module 4 is used to determine the target age information based on the identification information and send the target age information to the alarm detection module 5.

[0151] The alarm detection module 5 is used to determine whether the target age information meets the age requirement and execute the riding warning step according to the determination result.

[0152] The riding warning module 6 is used to execute the riding warning step.

[0153] The personnel riding monitoring and management system provided in this embodiment can implement the various steps of the aforementioned embodiments due to the functions of its modules themselves and the logical connections between each other, and can therefore achieve the same technical effects as the aforementioned embodiments. The principle analysis can be found in the relevant descriptions of the aforementioned method steps, which will not be repeated here.

[0154] Embodiment 4:

[0155] In one embodiment, a smart terminal is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the memory stores training data, algorithm formulas, and filtering mechanisms in a training model. The processor is used to provide computing and control capabilities, and when the processor executes the computer program, the following steps are implemented:

[0156] S1. Image acquisition: acquiring surveillance images.

[0157] S2, behavior detection: determine whether the monitoring image meets the detection conditions, and execute S3 according to the judgment result.

[0158] In step S2, it includes:

[0159] S21. Determine a human body area and a vehicle body area based on the monitoring image.

[0160] S22: Determine an overlapping area based on the human body area and the vehicle body area.

[0161] S23, determine whether the area of ​​the overlapping region is greater than the area threshold, if so, execute S3; if not, end.

[0162] S3. Information extraction: Determine identification information based on surveillance images.

[0163] In step S3, it includes:

[0164] S31. Determine skeleton information based on the monitoring image.

[0165] S32. Determine facial information based on the surveillance image.

[0166] In step S31, it includes:

[0167] S311, determining a target area image based on a human body area, a vehicle body area and a monitoring image.

[0168] S312: Perform image correction based on the target area image to determine an optimized image.

[0169] S313. Determine skeleton information based on the optimized image.

[0170] S4. Age estimation: Determine the target age information based on the identification information.

[0171] In step S4, it includes:

[0172] S41. Determine first age information based on the skeleton information.

[0173] S42: Determine second age information based on the facial information.

[0174] S43: Determine target age information based on the first age information and the second age information.

[0175] S5, alarm detection: determine whether the target age information meets the age limit condition, if yes, end; if not, execute S6.

[0176] S6. Riding warning.

[0177] In step S6, it includes:

[0178] S61. Send the facial information to the operator.

[0179] S62: The operator compares the facial information in the database to determine the identity information.

[0180] S63: The operator sends a warning message to the user terminal corresponding to the identity information.

[0181] The intelligent terminal provided in this embodiment can achieve the same technical effect as the aforementioned embodiment 2 because the computer program in its memory will implement the various steps of the aforementioned embodiment 2 after running on the processor. The principle analysis can refer to the relevant description of the aforementioned method steps, which will not be repeated here.

[0182] Embodiment five:

[0183] In one embodiment, a computer-readable storage medium is provided, which stores a computer program that can be loaded by a processor and execute the above-mentioned personnel riding monitoring management method. When the computer program is executed by the processor, the following steps are implemented:

[0184] S1. Image acquisition: acquiring surveillance images.

[0185] S2, behavior detection: determine whether the monitoring image meets the detection conditions, and execute S3 according to the judgment result.

[0186] In step S2, it includes:

[0187] S21. Determine a human body area and a vehicle body area based on the monitoring image.

[0188] S22: Determine an overlapping area based on the human body area and the vehicle body area.

[0189] S23, determine whether the area of ​​the overlapping region is greater than the area threshold, if so, execute S3; if not, end.

[0190] S3. Information extraction: Determine identification information based on surveillance images.

[0191] In step S3, it includes:

[0192] S31. Determine skeleton information based on the monitoring image.

[0193] S32. Determine facial information based on the surveillance image.

[0194] In step S31, it includes:

[0195] S311, determining a target area image based on a human body area, a vehicle body area and a monitoring image.

[0196] S312: Perform image correction based on the target area image to determine an optimized image.

[0197] S313. Determine skeleton information based on the optimized image.

[0198] S4. Age estimation: Determine the target age information based on the identification information.

[0199] In step S4, it includes:

[0200] S41. Determine first age information based on the skeleton information.

[0201] S42: Determine second age information based on the facial information.

[0202] S43: Determine target age information based on the first age information and the second age information.

[0203] S5, alarm detection: determine whether the target age information meets the age limit condition, if yes, end; if not, execute S6.

[0204] S6. Riding warning.

[0205] In step S6, it includes:

[0206] S61. Send the facial information to the operator.

[0207] S62: The operator compares the facial information in the database to determine the identity information.

[0208] S63: The operator sends a warning message to the user terminal corresponding to the identity information.

[0209] The readable storage medium provided in this embodiment can achieve the same technical effect as the aforementioned embodiment 2 because the computer program therein will implement the various steps of the aforementioned embodiment 2 after being loaded and run on the processor. The principle analysis can be found in the relevant description of the aforementioned method steps, which will not be repeated here.

[0210] The computer-readable storage medium includes, for example, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0211] The embodiments of the specific implementation methods are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the methods and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for monitoring and managing riding of personnel, characterized in that: include: Image acquisition: obtain surveillance images; Behavior detection: judging whether the monitoring image meets the detection condition, and executing the information extraction step according to the judgment result; wherein the detection condition is that the monitoring image records a vehicle and a rider riding on the vehicle; Information extraction: determining identification information based on the monitoring image; wherein the identification information includes skeleton information, and the skeleton information is used to reflect the body shape of the rider; Age estimation: Based on the identification information, determine the target age information; the age estimation step includes: Substitute the skeleton information into the age estimation model to determine the target age information; The skeleton information includes arm length information for reflecting the arm length of the rider, leg length information for reflecting the leg length of the rider, and spine information for reflecting the spine length of the rider; in the generation method of the age estimation model, it includes: Determine fitting information; wherein the fitting information includes age fitting information corresponding to people of different age groups, arm length fitting information corresponding to people of different age groups, leg length fitting information corresponding to people of different age groups, and spine fitting information corresponding to people of different age groups; the age fitting information is used to reflect the age group of the person; Substitute the fitting information into the multivariate function model for orthogonal polynomial fitting to generate an age estimation model; Alarm detection: Determine whether the target age information meets the age requirement, and execute the riding warning steps according to the judgment result.

2. The method for monitoring and managing riding of personnel according to claim 1, characterized in that: The specific method of the behavior detection step includes: Based on the monitoring image, determine a human body area and a vehicle body area; wherein the human body area is used to indicate the rider, and the vehicle body area is used to indicate the vehicle; Based on the human body area and the vehicle body area, determine the overlapping area; An area comparison is performed based on the overlapping area. If the area of ​​the overlapping area is greater than the area threshold, the monitoring image meets the detection condition and the information extraction step is executed.

3. The method for monitoring and managing riding of personnel according to claim 2, characterized in that: The specific method for determining identification information based on the surveillance image includes: Determine a target area image based on the human body area, the vehicle body area and the monitoring image; wherein the target area image can reflect the rider and the vehicle; Perform perspective transformation based on the target area image to determine the optimized image; Based on the optimized image, skeleton information is determined.

4. The method for monitoring and managing riding of personnel according to claim 1, characterized in that: The specific method of the information extraction step includes: Determine facial information based on surveillance images; Among the specific methods of age estimation are: Based on the identification information, determining first age information; Determining second age information based on the facial information; Based on the first age information and the second age information, target age information is determined.

5. The method for monitoring and managing riding of persons according to claim 4, characterized in that: The specific methods of riding warning steps include: Send facial information to the operator; The operator compares the facial information in the database to determine the identity information; The operator sends a warning message to the user terminal corresponding to the identity information.

6. The personnel riding monitoring and management system is characterized by: include: An image acquisition module (1) is used to acquire monitoring images; a behavior detection module (2), used to determine whether the monitoring image meets the detection condition, and to perform the information extraction step according to the determination result; wherein the detection condition is that the monitoring image records a vehicle and a rider riding on the vehicle; An information extraction module (3) is used to determine identification information based on the monitoring image, wherein the identification information includes skeleton information, and the skeleton information is used to reflect the body shape of the rider; The age estimation module (4) is used to determine target age information based on the identification information; substitute the skeleton information into the age estimation model to determine the target age information; wherein the skeleton information includes arm length information for reflecting the arm length of the rider, leg length information for reflecting the leg length of the rider, and spine information for reflecting the spine length of the rider. In the method for generating the age estimation model, the method includes: determining fitting information; wherein the fitting information includes age fitting information corresponding to people of different age groups, arm length fitting information corresponding to people of different age groups, leg length fitting information corresponding to people of different age groups, and spine fitting information corresponding to people of different age groups; the age fitting information is used to reflect the age group of the person; and substitute the fitting information into the multivariate function model for orthogonal polynomial fitting to generate the age estimation model.

7. An alarm detection module (5) is used to determine whether the target age information meets the age requirement and execute a riding warning step according to the determination result.

8. Intelligent terminal, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executes any one of the methods of claims 1 to 5.

9. A computer-readable storage medium, characterized in that A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 5.

Images