Urban optimization system based on cross-media intelligence

Abstract

The present invention discloses a city optimization system based on cross-media intelligence, including a data acquisition module, a pedestrian threat detection module, a prompt necessity analysis module and an output module. The pedestrian threat detection module is used to detect potential threats encountered by pedestrians when walking; the prompt necessity analysis module is used to analyze the necessity of prompting vehicles traveling in the lane through the detection results of the pedestrian threat detection module. The present invention analyzes the ratio of pedestrians avoiding splashing water when walking on the road bricks in the sidewalk of the monitored section through the pedestrian walking monitoring module, which effectively reflects the construction integrity of the sidewalk; at the same time, the density of vehicles passing through the current section is analyzed through the vehicle distance detection submodule, and pedestrians are promptly reminded in the section with high splashing risk, thereby improving the prediction ability of the splashing risk of the monitored section and improving the fault tolerance rate of pedestrians walking safely. The present invention has the characteristics of strong real-time monitoring ability and high humanization.

Description

Technical Field

[0001] The present invention relates to the technical field of road order management, and in particular to a city optimization system based on cross-media intelligence. Background Art

[0002] Every year during the rainy season, the bricks on the sidewalks on some sections of the road are seriously damaged, which brings inconvenience to citizens. On some sidewalks, there are many broken tiles. When walking on such sidewalks on rainy days, it is not only easy for citizens to fall and get injured, but it is also easy to get splashed with water when stepping on them on rainy days.

[0003] The main reason for water accumulation on the sidewalk tiles after rain is that the sidewalk tiles become loose or damaged after long-term use, and rainwater flows into the gaps to form puddles. In the prior art, urban road maintenance personnel regularly maintain the sidewalks and repair loose and damaged road tiles. However, the degree of damage of road tiles in different locations in the city is different. Regular inspections are difficult to achieve timely maintenance of various roads in the city to ensure a good experience for citizens when walking on the sidewalks. Especially in sections near industrial areas, the sidewalks are affected by the splashing of a large number of transport vehicles, and it is easier to accumulate water in the road tiles; at the same time, the leakage of industrial raw materials during transportation also makes the pollution level of water accumulation in regional roads more serious than other urban roads, which greatly increases the possibility and degree of harm to pedestrians affected by splashing water. Therefore, it is necessary to design a city optimization system based on cross-media intelligence with strong real-time monitoring capabilities and high humanization. Summary of the invention

[0004] The purpose of the present invention is to provide a city optimization system based on cross-media intelligence to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: an urban optimization system based on cross-media intelligence, including a data acquisition module, a pedestrian threat detection module, a prompt necessity analysis module and an output module, the data acquisition module is used to obtain relevant information of urban roads; the pedestrian threat detection module is used to detect potential threats encountered by pedestrians when walking; the prompt necessity analysis module is used to analyze the necessity of prompting vehicles traveling in the lane through the detection results of the pedestrian threat detection module; and the output module is used to prompt passing vehicles.

[0006] According to the above technical solution, the data acquisition module includes a sidewalk information acquisition module and a lane information acquisition module. The sidewalk information acquisition module is used to obtain relevant construction parameter information of the sidewalk; the lane information acquisition module is used to obtain relevant lane line information in the lane.

[0007] According to the above technical solution, the pedestrian threat detection module includes a pedestrian walking monitoring module, a water splash analysis module and a camera unit. The pedestrian walking monitoring module is used to monitor and extract the characteristics of pedestrians walking on the sidewalk; the water splash analysis module is used to analyze and predict whether pedestrians will be splashed by water; and the camera unit is used to obtain images on the lane and sidewalk.

[0008] According to the above technical solution, the water splash analysis module further includes a road section importance prediction submodule and a vehicle distance detection submodule. The road section importance prediction submodule is used to predict the importance of the monitored road section by vehicle distance; the vehicle distance detection submodule is used to detect the maximum average vehicle distance between vehicles in the monitored road section.

[0009] According to the above technical solution, the output module includes a maintenance reminder module and a splash reminder module. The maintenance reminder module is used to remind urban road maintenance personnel to perform timely maintenance on the roads to be maintained; the splash reminder module is used to remind pedestrians on sidewalks where splash threats may occur.

[0010] According to the above technical solution, the operation method of the city optimization system mainly includes the following steps:

[0011] Step S1: Connect the urban optimization system to the road monitoring system of the road section near the industrial zone, and shoot the sidewalks and lanes in the road section through the cameras set on both sides of the road. When a vehicle enters the monitored road section, a connection relationship is established between the intelligent systems of the front and rear vehicles in the lane of the monitored road section to exchange information;

[0012] Step S2: The pedestrian walking monitoring module extracts the characteristics of pedestrians walking on the sidewalk and counts the rate at which the sampled pedestrians avoid splashing road bricks while walking on the monitored road section;

[0013] Step S3: combining the lane area division information acquired by the lane information acquisition module and the vehicle driving picture information acquired by the shooting unit, evaluating the importance of the traffic hub of the current monitored section;

[0014] Step S4: The prompt necessity analysis module combines the scene of cars passing in the lane to predict whether the distance of the splashes caused by the pedestrians walking on the road bricks will have a significant impact on the pedestrians. If it is judged that there is a significant impact, the information of the road section whose damage degree exceeds the safety value is output to the urban road maintenance personnel through the maintenance prompt module;

[0015] Step S5: In the captured images, images with obvious water splashing are collected, and the differences in the splashing characteristics between the front and rear splashing of the road bricks are extracted and analyzed to obtain the concentrated distance Y1 and the farthest distance Y2 of the splashing. The dangerous distance for pedestrians to be splashed is H=Y1+λY2, where G1 is the limited separation value between the vehicle distance during the peak driving period and the ordinary vehicle driving period calculated by the system, and λ is the distance elongation coefficient caused by the farthest distance of water splashing on the concentrated distance of water splashing, 0<λ<0.2. Road maintenance personnel record the dangerous distances of water splashing in sections at different locations in the city and arrange maintenance priorities according to the size of the water splashing distance.

[0016] According to the above technical solution, step S2 further includes:

[0017] Step S21: The pedestrian walking monitoring module monitors the pedestrians moving in the shooting picture, and monitors the minimum number of times the target pedestrians are expected to step on the monitored road section. Where G is the average step length of a pedestrian, and the length of the monitored section is L;

[0018] Step S22: When a pedestrian appears in the shooting picture of the pedestrian road brick section, the distribution of the road bricks on the sidewalk is used as a reference for the road trajectory. When the pedestrian enters the shooting picture, the actual stepping position of the pedestrian on the sidewalk road bricks and the pedestrian's direction are recorded to predict the pedestrian's walking trajectory when walking in a straight line, and the overlap between the actual stepping position of the pedestrian's foot and the predicted walking trajectory is monitored in real time;

[0019] Step S23: Obtain the number of records of the three situations where the actual stepping position of the pedestrian entering the shooting screen is within the walking track, the actual stepping position is at the edge of the walking track graph, and the actual stepping position is outside the walking track, respectively. The ratio of pedestrians avoiding splashing road bricks while walking in the monitoring section is A, B, and C.

[0020] Step S24: When When the pedestrian sampling sidewalk section is determined to be a section to be maintained; otherwise, the pedestrian walking sampling sidewalk section is determined to be a normal section; during the monitoring time, if the number of "sections to be maintained" obtained by section sampling is higher than the number of "normal sections", the prompt information will be given through the roadside signal signs.

[0021] According to the above technical solution, in step S22, the actual stepping position of the pedestrian on the sidewalk bricks is recorded in the monitoring screen of the camera as a circle with a radius of 1dm based on the largest area of ​​the pedestrian's sole, and is calibrated in real time in the image of the monitoring screen.

[0022] According to the above technical solution, step S3 further includes:

[0023] Step S31: The cameras set up on both sides of the road section capture and record the driving images of vehicles in a single lane closest to the sidewalk. The number of vehicles in the monitored road section during the time period is analyzed every time period K. The number of vehicles that fully enter the monitored road section per second is X1, X2, ..., X. K , the average length of ordinary cars is S, and there are P cars; the average length of transport trucks is ηS, and there are Q cars, where η is the length increase ratio of trucks purchased in bulk relative to ordinary cars according to the current industrial zone records, The maximum average distance between vehicles during the monitoring period is Among them G max The unit is meter;

[0024] Step S32: The distance limit value between two vehicles on the current monitoring road G=N+V max T, the importance of the traffic hub of the current road section Where N is the safe distance between the two vehicles, T is the normal reaction time of the driver, V max The maximum speed limit for the current road, with G in meters.

[0025] According to the above technical solution, in step S4, according to the splashing frequency f of the sidewalk bricks in the monitored section obtained in step S23 and the traffic hub importance η of the monitored section obtained in step S32, the maintenance necessity rate of the current section is specifically:

[0026]

[0027] When φ>φ0, the information of the road section where the degree of damage exceeds the safety value is output to the urban road maintenance personnel through the maintenance reminder module, where φ0 is the limit ratio of the damage degree of the sidewalk paving bricks obtained through the database to the maintenance standard.

[0028] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention, through the pedestrian walking monitoring module, analyzes the ratio of pedestrians avoiding splashing water when walking on paving bricks in the sidewalk of the monitored section, effectively reflects the construction integrity of the sidewalk, and increases the accuracy of monitoring while improving the wide application of the system; at the same time, the vehicle distance detection submodule analyzes the density of vehicles passing through the current section, and promptly reminds pedestrians in sections with high splashing risks, thereby improving the predictive ability of splashing risks in the monitored section and improving the fault tolerance rate of pedestrians walking safely; finally, the prompt necessity analysis module is combined with the concentrated distance and the farthest distance of splashing water obtained by the system and the risk distance is calculated, which greatly reduces the misjudgment of pedestrian walking, improves the safety of pedestrian walking, and increases the accuracy of system detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0030] Figure 1 It is a schematic diagram of the system module composition of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] Please refer to Figure 1 The present invention provides a technical solution: a city optimization system based on cross-media intelligence, comprising:

[0033] The data acquisition module, pedestrian threat detection module, prompt necessity analysis module and output module are used. The data acquisition module is used to obtain relevant information of urban roads; the pedestrian threat detection module is used to detect potential threats encountered by pedestrians when walking; the prompt necessity analysis module is used to analyze the necessity of prompting vehicles traveling in the lane through the detection results of the pedestrian threat detection module; and the output module is used to prompt passing vehicles.

[0034] The present invention uses a pedestrian walking monitoring module to analyze the ratio of pedestrians avoiding splashing water when walking on paving bricks in the sidewalk of the monitored section, which effectively reflects the construction integrity of the sidewalk, and improves the accuracy of monitoring while improving the wide application of the system; at the same time, the vehicle distance detection submodule is used to analyze the density of vehicles passing through the current section, and pedestrians are reminded in time in sections with high splashing risks, thereby improving the prediction ability of splashing risks in the monitored section and improving the fault tolerance rate of pedestrians walking safely; finally, the prompt necessity analysis module is combined with the concentrated distance and the farthest distance of splashing water obtained by the system and the risk distance is calculated, which greatly reduces the misjudgment of pedestrian walking, improves the safety of pedestrian walking, and increases the accuracy of system detection.

[0035] The data acquisition module includes a sidewalk information acquisition module and a lane information acquisition module. The sidewalk information acquisition module is used to obtain relevant construction parameter information of the sidewalk; the lane information acquisition module is used to obtain relevant lane line information in the lane.

[0036] The pedestrian threat detection module includes a pedestrian walking monitoring module, a water splash analysis module and a camera unit. The pedestrian walking monitoring module is used to monitor and extract the characteristics of pedestrians walking on the sidewalk; the water splash analysis module is used to analyze and predict whether pedestrians will be splashed by water; and the camera unit is used to obtain images on the lanes and sidewalks.

[0037] The water splash analysis module further includes a road section importance prediction submodule and a vehicle distance detection submodule. The road section importance prediction submodule is used to predict the importance of the monitored road section through the vehicle distance; the vehicle distance detection submodule is used to detect the maximum average vehicle distance between vehicles in the monitored road section.

[0038] The output module includes a maintenance reminder module and a splash reminder module. The maintenance reminder module is used to remind urban road maintenance personnel to carry out timely maintenance on roads to be maintained; the splash reminder module is used to remind pedestrians on sidewalks where splash threats may occur.

[0039] In a preferred embodiment, the operation method of the city optimization system mainly includes the following steps:

[0040] Step S1: Connect the urban optimization system to the road monitoring system of the road section near the industrial zone, and shoot the sidewalks and lanes in the road section through the cameras set on both sides of the road. When a vehicle enters the monitored road section, a connection relationship is established between the intelligent systems of the front and rear vehicles in the lane of the monitored road section to exchange information;

[0041] Step S2: The pedestrian walking monitoring module extracts the characteristics of pedestrians walking on the sidewalk and counts the rate at which the sampled pedestrians avoid splashing road bricks while walking on the monitored road section;

[0042] Step S3: combining the lane area division information acquired by the lane information acquisition module and the vehicle driving picture information acquired by the shooting unit, evaluating the importance of the traffic hub of the current monitored section;

[0043] Step S4: The prompt necessity analysis module combines the scene of cars passing in the lane to predict whether the distance of the splashes caused by the pedestrians walking on the road bricks will have a significant impact on the pedestrians. If it is judged that there is a significant impact, the information of the road section whose damage degree exceeds the safety value is output to the urban road maintenance personnel through the maintenance prompt module;

[0044] Step S5: In the captured images, images with obvious water splashing are collected, and the differences in the splashing characteristics between the front and rear splashing of the road bricks are extracted and analyzed to obtain the concentrated distance Y1 and the farthest distance Y2 of the splashing. The dangerous distance for pedestrians to be splashed is H=Y1+λY2, where G1 is the limited separation value between the vehicle distance during the peak driving period and the ordinary vehicle driving period calculated by the system, and λ is the distance elongation coefficient caused by the farthest distance of water splashing on the concentrated distance of water splashing, 0<λ<0.2. Road maintenance personnel record the dangerous distances of water splashing in sections at different locations in the city and arrange maintenance priorities according to the size of the water splashing distance.

[0045] The necessity of maintenance is that the target road needs to be repaired after the standard is met, but the necessity rate of maintenance of each road cannot directly reflect the maintenance priority of the road. By detecting the splash distance, the walking experience of pedestrians can be predicted and the road maintenance priority can be evaluated accordingly.

[0046] There are misunderstandings in the calculation of the distance of water splashing in the existing technology. The distance at which pedestrians perceive water splashing is much lower than the farthest distance of water splashing. This technical solution combines the concentrated distance and the farthest distance of water splashing obtained by the system and calculates the risk distance, which greatly reduces the misjudgment of water splashing while pedestrians are walking, provides a strong basis for maintenance personnel to maintain roads, improves pedestrian walking safety, and increases the accuracy of system detection.

[0047] In this embodiment, step S2 further includes:

[0048] Step S21: The pedestrian walking monitoring module monitors the pedestrians moving in the shooting picture, and monitors the minimum number of times the target pedestrians are expected to step on the monitored road section. Where G is the average step length of a pedestrian, the length of the monitored section is L, and within the section captured by a single monitoring camera, the monitoring distances of the lane and the sidewalk are the same;

[0049] Step S22: When a pedestrian appears in the shooting picture of the pedestrian road brick section, the distribution of the road bricks on the sidewalk is used as a reference for the road trajectory. When the pedestrian enters the shooting picture, the actual stepping position of the pedestrian on the sidewalk road bricks and the pedestrian's direction are recorded to predict the pedestrian's walking trajectory when walking in a straight line, and the overlap between the actual stepping position of the pedestrian's foot and the predicted walking trajectory is monitored in real time. The walking trajectory is specifically a virtual rectangular area in the shooting section;

[0050] Step S23: Obtain the number of records of the three situations where the actual stepping position of the pedestrian entering the shooting screen is within the walking track, the actual stepping position is at the edge of the walking track graph, and the actual stepping position is outside the walking track, respectively. The ratio of pedestrians avoiding splashing road bricks while walking in the monitoring section is A, B, and C.

[0051] Step S24: When When the pedestrian sampling sidewalk section is determined to be a section to be maintained; otherwise, the pedestrian walking sampling sidewalk section is determined to be a normal section; during the monitoring time, if the number of "sections to be maintained" obtained by section sampling is higher than the number of "normal sections", the prompt information will be given through the roadside signal signs.

[0052] The higher the road construction degree, the higher the integrity of the pedestrian pavement bricks on both sides of the road, the less likely pedestrians are to be splashed by the pavement bricks while walking, and the faster the average speed of pedestrians walking on rainy days; conversely, the lower the road construction degree, the lower the integrity of the pedestrian pavement bricks on both sides of the road, the greater the possibility of pedestrians being splashed by the pavement bricks while walking, and the slower the average speed of pedestrians walking on rainy days.

[0053] In step S22 of this embodiment, the actual stepping position of the pedestrian on the sidewalk brick is recorded in the monitoring image of the camera as a circle with a radius of 1 dm based on the largest area of ​​the pedestrian's sole, and is calibrated in real time in the image of the monitoring image.

[0054] In this embodiment, step S3 further includes:

[0055] Step S31: The cameras set up on both sides of the road section capture and record the driving images of vehicles in a single lane closest to the sidewalk. The number of vehicles in the monitored road section during the time period is analyzed every time period K. The number of vehicles that fully enter the monitored road section per second is X1, X2, ..., X. K , the average length of ordinary cars is S, and there are P cars; the average length of transport trucks is ηS, and there are Q cars, where η is the length increase ratio of trucks purchased in bulk relative to ordinary cars according to the current industrial zone records, The maximum average distance between vehicles during the monitoring period is Among them G max The unit is meter;

[0056] During the monitoring period, the system counts the average length of vehicles passing through the monitored road section; The vehicle distance predicted by the monitoring screen obtained by the system at the i-th second in the monitoring time period. Since the camera is set at a high position and the shooting angle is not perpendicular to the vehicle's driving direction, and the distance between vehicles is constantly changing, it is difficult for a fixed camera to directly obtain the average distance between vehicles in the lane over a period of time through monitoring;

[0057] Step S32: The distance limit value between two vehicles on the current monitoring road G=N+V max T, the importance of the traffic hub of the current road section Where N is the safe distance between the two vehicles, T is the normal reaction time of the driver, V max The maximum speed limit for the current road, with G in meters.

[0058] In the monitored section, since the sections are built uniformly during road construction, the road integrity of the lanes and sidewalks in the section is relatively uniform. By analyzing the average vehicle distance in the lanes of the monitored section, it is indirectly reflected the frequency of vehicles passing through the road, and the current water accumulation on the pavement caused by vehicle splashing can be judged based on the vehicle driving frequency. While increasing the accuracy of monitoring, it can be applicable to roads of various completeness, thereby improving the wide application of the system.

[0059] In step S4 of this embodiment, according to the splashing frequency f of the sidewalk bricks in the monitoring section obtained in step S23 and the traffic hub importance η of the monitoring section obtained in step S32, the maintenance necessity rate of the current section is specifically:

[0060]

[0061] When φ>φ0, the system assesses that the pedestrians on the current sidewalk are obviously affected by the splashing water, and outputs the information of the road section where the degree of damage exceeds the safety value to the urban road maintenance personnel through the maintenance reminder module, where φ0 is the limit ratio of the sidewalk paving brick damage degree reaching the maintenance standard obtained through the database.

[0062] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0063] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. The city optimization system based on cross-media intelligence is characterized by: It includes a data acquisition module, a pedestrian threat detection module, a prompt necessity analysis module and an output module. The data acquisition module is used to obtain relevant information of urban roads; the pedestrian threat detection module is used to detect potential threats encountered by pedestrians when walking; the prompt necessity analysis module is used to analyze the necessity of prompting vehicles traveling in the lane through the detection results of the pedestrian threat detection module; and the output module is used to prompt passing vehicles; The data acquisition module includes a sidewalk information acquisition module and a lane information acquisition module, and the sidewalk information acquisition module is used to obtain relevant construction parameter information of the sidewalk; The lane information acquisition module is used to obtain relevant lane line information within the lane; The pedestrian threat detection module includes a pedestrian walking monitoring module, a water splashing analysis module and a camera unit. The pedestrian walking monitoring module is used to monitor and extract the characteristics of pedestrians walking on the sidewalk; the water splashing analysis module is used to analyze and predict whether pedestrians will be splashed by water; and the camera unit is used to obtain images on the lane and the sidewalk; The water splash analysis module further includes a road section importance prediction submodule and a vehicle distance detection submodule, wherein the road section importance prediction submodule is used to predict the importance of the monitored road section by vehicle distance; and the vehicle distance detection submodule is used to detect the maximum average vehicle distance between vehicles in the monitored road section; The output module includes a maintenance reminder module and a water splash reminder module, wherein the maintenance reminder module is used to remind urban road maintenance personnel to perform timely maintenance on the roads to be maintained; The water splashing warning module is used to warn pedestrians on the sidewalk where the system detects a water splashing threat; The operation method of the city optimization system comprises the following steps: Step S1: Connect the urban optimization system to the road monitoring system of the road section near the industrial zone, and shoot the sidewalks and lanes in the road section through the cameras set on both sides of the road. When a vehicle enters the monitored road section, a connection relationship is established between the intelligent systems of the front and rear vehicles in the lane of the monitored road section to exchange information; Step S2: The pedestrian walking monitoring module extracts the characteristics of pedestrians walking on the sidewalk and counts the rate at which the sampled pedestrians avoid splashing road bricks while walking on the monitored road section; Step S3: combining the lane area division information acquired by the lane information acquisition module and the vehicle driving picture information acquired by the shooting unit, evaluating the importance of the traffic hub of the current monitored section; Step S4: The prompt necessity analysis module combines the scene of cars passing in the lane to predict whether the distance of the splashes caused by the pedestrians walking on the road bricks will have a significant impact on the pedestrians. If it is judged that there is a significant impact, the information of the road section whose damage degree exceeds the safety value is output to the urban road maintenance personnel through the maintenance prompt module; Step S5: In the captured images, images with obvious water splashing are collected, and the differences in the splashing characteristics between the front and rear splashing of the road bricks are extracted and analyzed to obtain the concentrated distance Y1 and the farthest distance Y2 of the splashing. The dangerous distance for pedestrians to be splashed is H=Y1+λY2, where G1 is the limited separation value between the vehicle distance during the peak driving period and the ordinary vehicle driving period calculated by the system, and λ is the distance elongation coefficient caused by the farthest distance of water splashing on the concentrated distance of water splashing, 0<λ<0.

2. Road maintenance personnel record the dangerous distances of water splashing in sections at different locations in the city and arrange maintenance priorities according to the size of the water splashing distance.

2. The city optimization system based on cross-media intelligence according to claim 1 is characterized in that: The step S2 further comprises: Step S21: The pedestrian walking monitoring module monitors the pedestrians moving in the shooting picture, and monitors the minimum number of times the target pedestrians are expected to step on the monitored road section. Where G is the average step length of a pedestrian, and the length of the monitored section is L; Step S22: When a pedestrian appears in the shooting picture of the pedestrian road brick section, the distribution of the road bricks on the sidewalk is used as a reference for the road trajectory. When the pedestrian enters the shooting picture, the actual stepping position of the pedestrian on the sidewalk road bricks and the pedestrian's direction are recorded to predict the pedestrian's walking trajectory when walking in a straight line, and the overlap between the actual stepping position of the pedestrian's foot and the predicted walking trajectory is monitored in real time; Step S23: Obtain the number of records of the three situations where the actual stepping position of the pedestrian entering the shooting screen is within the walking track, the actual stepping position is at the edge of the walking track graph, and the actual stepping position is outside the walking track, respectively. The ratio of pedestrians avoiding splashing road bricks while walking in the monitoring section is A, B, and C. Step S24: When When the pedestrian sampling sidewalk section is determined to be a section to be maintained; otherwise, the pedestrian walking sampling sidewalk section is determined to be a normal section; during the monitoring time, if the number of "sections to be maintained" obtained by section sampling is higher than the number of "normal sections", the prompt information will be given through the roadside signal signs.

3. The city optimization system based on cross-media intelligence according to claim 2 is characterized in that: In step S22, the actual stepping position of the pedestrian on the sidewalk brick is recorded in the monitoring image of the camera as a circle with a radius of 1 dm based on the largest area of ​​the pedestrian's sole, and is calibrated in real time in the image of the monitoring image.

4. The city optimization system based on cross-media intelligence according to claim 3 is characterized in that: The step S3 further comprises: Step S31: The cameras set up on both sides of the road section capture and record the driving images of vehicles in a single lane closest to the sidewalk. The number of vehicles in the monitored road section during the time period is analyzed every time period K. The number of vehicles that fully enter the monitored road section per second is X1, X2, ..., X. K , the average length of ordinary cars is S, and there are P cars; the average length of transport trucks is δS, and there are Q cars, where δ is the length increase ratio of trucks purchased in batches relative to ordinary cars according to the current industrial zone records, The maximum average distance between vehicles during the monitoring period is Among them G max The unit is meter; Step S32: The distance limit value between two vehicles on the current monitoring road G=N+V max T, the importance of the current road segment Where N is the safe distance between the two vehicles, T is the normal reaction time of the driver, V max The maximum speed limit for the current road, with G in meters.

5. The city optimization system based on cross-media intelligence according to claim 4 is characterized in that: In step S4, according to the ratio f of pedestrians avoiding splashed road bricks while walking in the monitored area obtained in step S23 and the importance η of the current road section obtained in step S32, the maintenance necessity rate of the current road section is specifically: When φ>φ0, the information of the road section where the degree of damage exceeds the safety value is output to the urban road maintenance personnel through the maintenance reminder module, where φ0 is the limit ratio of the damage degree of the sidewalk paving bricks obtained through the database to the maintenance standard.

Images