Bus stop capacity prediction method based on video analysis

Abstract

The present invention discloses a method for predicting bus stop capacity based on video analysis, comprising the following steps: first data acquisition, second data acquisition, data calculation, third data acquisition, comprehensive data acquisition, comprehensive data analysis, and comprehensive data comparison. The present invention relates to the field of road technology. By obtaining the intersection capacity during peak hours, the time impact value, and the intersection capacity at the time of a traffic accident, the present invention effectively reflects the factors influencing bus stop capacity, accurately calculating and predicting bus stop capacity. Furthermore, by obtaining the average of the difference in the capacity and number of stations on a designated road and its adjacent roads, the present invention obtains a preferred factor for bus stop construction, calculates the capacity of unbuilt bus stops, and determines the optimal location for the bus stop. This prevents a decline in service levels caused by irrational bus stop construction and reduces one-way travel time for the public transportation system.

Description

Technical Field

[0001] The present invention relates to the field of road technology, and in particular to a method for predicting bus stop capacity based on video analysis. Background Art

[0002] The capacity of a bus stop refers to the maximum number of buses that a bus stop can accommodate, taking into account the actual road and traffic conditions. To improve the capacity of a bus stop, the existing method is to generally increase the number of stops at the stop. This method can reduce the stop time of "last-in-first-out" buses. However, some roads are not suitable for increasing the number of stops at the stop. For bus stops with poor road capacity, the "last-in-first-out" method of buses is very likely to cause congestion of private vehicles on the road. This plan uses the impact of signal control on the capacity of the bus stop to effectively analyze the site selection before the construction of the bus stop, provide a guarantee for correctly estimating the transportation capacity of the entire bus lane system, and provide a basis for the reasonable planning, construction and management of bus lanes. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the present invention provides a bus stop capacity prediction method based on video analysis, which solves the problems raised in the background technology.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: a bus stop capacity prediction method based on video analysis, comprising the following steps:

[0005] Step 1: First data acquisition

[0006] On a designated road, the nearest intersection is determined based on the planned travel direction. The intersection capacity Gi during peak hours is calculated using the traffic light cycle, the green light time for each direction within a traffic light cycle, and the number of vehicles in each direction.

[0007] Step 2: Acquisition of the Second Data

[0008] In one cycle, all traffic accident videos on a designated road are collected, and the average vehicle speed before and after the accident is obtained. Then, the time impact value BS and the intersection capacity SGi at the time of the accident are obtained based on the accident duration.

[0009] Step 3: Data calculation

[0010] The platform capacity GN is obtained through the formula GN=Gi+SGi*BS.

[0011] Preferably, the specific calculation method of the fork capacity Gi in step 1 is as follows:

[0012] S1. During peak hours, collect the traffic light cycle Tz of a specified road and the green light time Tdi of each driving direction within a traffic light cycle, where i = 1, 2, or 3, and the driving directions are the straight, left, and right directions on the traffic road, with 1 representing the straight direction, 2 representing the left turn, and 3 representing the right turn;

[0013] The signal light cycle is the total time it takes for all lights in a traffic light to appear in sequence.

[0014] S2. Obtain the number of vehicles Ci in each driving direction within a traffic light cycle through road monitoring video. The number of vehicles passing through the road stop line when the traffic light is green is counted to obtain the vehicle number.

[0015] Then, the average time for vehicles to pass the road stop line in each driving direction is obtained by Ti = Tdi / Ci;

[0016] S3. Obtain the intersection capacity Gi for each driving direction of the designated road through the formula Gi=3600 / Tz*{(Tdi-Ty) / Ti+1}*β, where Ty is a preset fixed value and β is a reduction coefficient, which is 0.88.

[0017] Preferably, all traffic accident videos are queried through a terminal in the office hall of the local traffic brigade, and the average vehicle speed is the average value of multiple vehicle speeds detected by a fixed-point test electronic eye.

[0018] Preferably, the specific calculation method of the fork junction capacity SGi in step 2 is:

[0019] AS1. Taking a set of traffic accident videos as an example, the average vehicle speed before the traffic accident is used as the reference speed Vc. Then, the average vehicle speed Vi before and after the traffic accident is obtained at regular intervals, where i = 1, 2, 3, ...;

[0020] Let i = 1, and compare V1 with Vc. If V1 ≥ Vc, then V1 and Vc will not be compared in the next step.

[0021] If V1 < Vc, then let i = 2, 3, ..., and compare Vi with Vc until Vi ≥ Vc, then V1 and Vc stop the next comparison, indicating that the traffic accident has been cleared;

[0022] AS2. Afterwards, obtain the duration from the start of the traffic accident to the time it is cleared, and record it as the accident duration;

[0023] Similarly, the accident duration of all traffic accidents is obtained and marked as Sj, where j = 1, 2, 3, ..., n, indicating that the number of accidents is n;

[0024] AS3. Mark the duration of the period as Sz and obtain the time impact value BS using the formula BS = (S1 + S2 + ..., + Sn) / Sz;

[0025] AS4. Then, according to the specific calculation method of the fork capacity during the peak period, the fork capacity during the accident period is calculated and marked as SGi.

[0026] Preferably, the bus stop capacity prediction method based on video analysis further includes the following steps after steps one to three:

[0027] Step 4: Acquisition of the third data

[0028] For the shared bicycle parking spaces planned on designated roads, the average difference in the number of shared bicycles before and after rush hour is calculated;

[0029] Step 5: Comprehensive data acquisition

[0030] The principles are the same as those of steps 1 to 4. The average value of the platform capacity and number difference of other roads adjacent to the designated road is obtained.

[0031] Step 6: Comprehensive data analysis

[0032] The average value of the difference between the number of the designated road and other roads is marked as EPr, and the platform capacity of the designated road and other roads is GNr, r = 1, 2, ... v, indicating that the total number of roads of the designated road and other roads is v;

[0033] Then, the optimization factor for bus stop construction is obtained by Yr = {EPr / (EP1+EP2+, ..., +EPv)} + GNr. The optimization factor for bus stop construction is the evaluation value of the best location for bus stops among multiple roads.

[0034] Step 7: Comprehensive data comparison

[0035] Then, the preferred factors of the bus stop construction on the designated road are compared with those on other roads, and the road with the largest preferred factor value is obtained as the preferred construction road for the bus stop.

[0036] Preferably, the specific calculation method of the average value of the difference in the number of shared bicycles in step 4 is:

[0037] S1. Through questionnaire surveys and interviews, the rush hour is divided into the front and back periods, and the front and back periods are obtained;

[0038] S2. Collect the number of shared bicycles in the first and second time periods respectively, and then calculate the absolute value of the difference between the number of shared bicycles in the first and second time periods as the number difference. The number of shared bicycles dispatched out and / or transferred in by the shared bicycle dispatcher in the first and second time periods is not included in the calculation;

[0039] Similarly, in different time periods, multiple groups of quantity differences are obtained and marked as Ek, where k = 1, 2, ..., m, indicating that there are m groups of quantity differences;

[0040] S3. Calculate the average value EP of the quantity differences through EP=(E1+E2+, ..., +Em) / m.

[0041] Beneficial effects

[0042] The present invention provides a method for predicting bus stop capacity based on video analysis. Compared with the existing technology, it has the following advantages:

[0043] The present invention effectively reflects the factors affecting bus stop capacity by obtaining the fork capacity, time impact value and fork capacity during peak hours when a traffic accident occurs. Therefore, the present invention can accurately calculate and predict bus stop capacity. It also obtains the average value of the platform capacity and quantity difference of a specified road and other adjacent roads to obtain the optimal factor for bus stop construction, further playing the following role: the capacity of unbuilt bus stops can be calculated, and then the optimal location of bus stops on multiple roads can be determined, thereby avoiding the decline in service level caused by unreasonable construction of bus stops and further reducing the one-way travel time of the public transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0046] See also Figure 1 The present invention provides a technical solution: a bus stop capacity prediction method based on video analysis, as a first embodiment of the present invention, comprising the following steps:

[0047] Step 1: First data acquisition

[0048] On a designated road, the nearest intersection is determined based on the planned travel direction. The intersection capacity Gi during peak hours is calculated using the traffic light cycle, the green light time for each direction within a traffic light cycle, and the number of vehicles in each direction.

[0049] The specific calculation method for the intersection capacity during peak hours is as follows:

[0050] S1. During peak hours, collect the traffic light cycle Tz of a specified road and the green light time Tdi of each driving direction within a traffic light cycle, where i = 1, 2, or 3, and the driving directions are the straight, left, and right directions on the traffic road, with 1 representing the straight direction, 2 representing the left turn, and 3 representing the right turn;

[0051] The signal light cycle is the total time it takes for all lights in a traffic light to appear in sequence.

[0052] S2. Obtain the number of vehicles Ci in each driving direction within a traffic light cycle through road monitoring video. The number of vehicles passing through the road stop line when the traffic light is green is counted to obtain the vehicle number.

[0053] Then, the average time for vehicles to pass the road stop line in each driving direction is obtained by Ti = Tdi / Ci;

[0054] S3. Obtain the intersection capacity Gi for each direction of travel of the designated road using the formula Gi = 3600 / Tz*{(Tdi-Ty) / Ti+1}*β, where Ty is a preset fixed value and β is a reduction factor, which is 0.88;

[0055] Step 2: Acquisition of the Second Data

[0056] In one cycle, all traffic accident videos on a designated road are collected, and the average vehicle speed before and after the accident is obtained. Then, the time impact value BS and the intersection capacity SGi at the time of the accident are obtained based on the accident duration.

[0057] All traffic accident videos are queried through the terminal in the office hall of the local traffic brigade. The average vehicle speed is the average of the speeds of multiple vehicles detected by the fixed-point test electronic eye.

[0058] The specific calculation method of the time impact value and the intersection capacity when a traffic accident occurs is:

[0059] AS1. Taking a set of traffic accident videos as an example, the average vehicle speed before the traffic accident is used as the reference speed Vc. Then, the average vehicle speed Vi before and after the traffic accident is obtained at regular intervals, where i = 1, 2, 3, ...;

[0060] Let i = 1, and compare V1 with Vc. If V1 ≥ Vc, then V1 and Vc will not be compared in the next step.

[0061] If V1 < Vc, then let i = 2, 3, ..., and compare Vi with Vc until Vi ≥ Vc, then V1 and Vc stop the next comparison, indicating that the traffic accident has been cleared;

[0062] AS2. Afterwards, obtain the duration from the start of the traffic accident to the time it is cleared, and record it as the accident duration;

[0063] Similarly, the accident duration of all traffic accidents is obtained and marked as Sj, where j = 1, 2, 3, ..., n, indicating that the number of accidents is n;

[0064] AS3. Mark the duration of the period as Sz and obtain the time impact value BS using the formula BS = (S1 + S2 + ..., + Sn) / Sz;

[0065] AS4 then calculates the capacity of the intersection during the accident period according to the specific calculation method of the intersection capacity during the peak period and marks it as SGi;

[0066] Step 3: Data calculation

[0067] The platform capacity GN is obtained by the formula GN = Gi + SGi * BS;

[0068] As the second embodiment of the present invention, this embodiment further adds the following steps based on the first embodiment:

[0069] Step 4: Acquisition of the third data

[0070] In the shared bicycle parking spaces planned on the designated roads, the average difference in the number of shared bicycles before and after the rush hour is calculated. The specific calculation method is:

[0071] S1. Through questionnaire surveys and interviews, the rush hour is divided into the front and back periods, and the front and back periods are obtained;

[0072] S2. Collect the number of shared bicycles in the first and second time periods respectively, and then calculate the absolute value of the difference between the number of shared bicycles in the first and second time periods as the number difference. The number of shared bicycles dispatched out and / or transferred in by the shared bicycle dispatcher in the first and second time periods is not included in the calculation;

[0073] Similarly, in different time periods, multiple groups of quantity differences are obtained and marked as Ek, where k = 1, 2, ..., m, indicating that there are m groups of quantity differences;

[0074] S3. Calculate the average value EP of the quantity differences by EP = (E1 + E2 + ..., + Em) / m;

[0075] Step 5: Comprehensive data acquisition

[0076] The principles are the same as those of steps 1 to 4. The average value of the platform capacity and number difference of other roads adjacent to the designated road is obtained.

[0077] Step 6: Comprehensive data analysis

[0078] The average value of the difference between the number of the designated road and other roads is marked as EPr, and the platform capacity of the designated road and other roads is GNr, r = 1, 2, ... v, indicating that the total number of roads of the designated road and other roads is v;

[0079] Then, the optimization factor for bus stop construction is obtained by Yr = {EPr / (EP1+EP2+, ..., +EPv)} + GNr. The optimization factor for bus stop construction is the evaluation value of the best location for bus stops among multiple roads.

[0080] Step 7: Comprehensive data comparison

[0081] Then, the preferred factors of the bus stop construction on the designated road are compared with those on other roads, and the road with the largest preferred factor value is obtained as the preferred construction road for the bus stop;

[0082] As the third embodiment of the present invention, this embodiment integrates the first and second embodiments;

[0083] By obtaining the fork capacity, time impact value and fork capacity during peak hours when traffic accidents occur, the factors affecting the bus stop capacity are effectively reflected. Therefore, the present invention can accurately calculate and predict the bus stop capacity. By obtaining the average value of the station capacity and quantity difference of the specified road and other adjacent roads, the optimal factor for bus stop construction is obtained, which further plays the following role: the capacity of unbuilt bus stops can be calculated, and then the optimal location of bus stops on multiple roads can be determined, thereby avoiding the decline in service level caused by unreasonable construction of bus stops and further reducing the one-way travel time of the public transportation system.

[0084] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0085] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A bus stop capacity prediction method based on video analysis is characterized by: The following steps are involved: Step 1: First data acquisition On a designated road, the nearest intersection is determined based on the planned travel direction. The intersection capacity Gi during peak hours is calculated using the traffic light cycle, the green light time for each direction within a traffic light cycle, and the number of vehicles in each direction. Step 2: Acquisition of the Second Data In one cycle, all traffic accident videos on a designated road are collected, and the average vehicle speed before and after the accident is obtained. Then, the time impact value BS and the intersection capacity SGi at the time of the accident are obtained based on the accident duration. Step 3: Data calculation The platform capacity GN is obtained through the formula GN=Gi+SGi*BS; The specific calculation method of the fork capacity SGi in step 2 is: AS1. Taking a set of traffic accident videos as an example, the average vehicle speed before the traffic accident is used as the reference speed Vc. Then, the average vehicle speed Vi before and after the traffic accident is obtained at regular intervals, where i = 1, 2, 3, ...; Let i = 1, and compare V1 with Vc. If V1 ≥ Vc, then V1 and Vc will not be compared in the next step. If V1 < Vc, then let i = 2, 3, ..., and compare V1 with Vc until Vi ≥ Vc, then V1 and Vc stop the next comparison, indicating that the traffic accident has been cleared; AS2. Afterwards, obtain the duration from the start of the traffic accident to the time it is cleared, and record it as the accident duration; By analogy, the accident duration of all traffic accidents is obtained and marked as Sj, where j = 1, 2, 3, ..., n, indicating that the number of accidents is n; AS3. Mark the duration of the period as Sz and obtain the time impact value BS using the formula BS = (S1 + S2 + ..., + Sn) / Sz; AS4. Then, according to the specific calculation method of the fork capacity during the peak period, the fork capacity during the accident period is calculated and marked as SGi.

2. The method for predicting bus stop capacity based on video analysis according to claim 1 is characterized in that: The specific calculation method of the fork capacity Gi in step 1 is as follows: S1. During peak hours, collect the traffic light cycle Tz of a designated road and the green light time Tdi of each driving direction within a traffic light cycle, where i = 1, 2, or 3, and the driving direction is the straight direction, left turn direction, and right turn direction on the traffic road, with 1 representing the straight direction, 2 representing the left turn direction, and 3 representing the right turn direction; The signal light cycle is the total time it takes for all lights in a traffic light to appear in sequence. S2. Obtain the number of vehicles Ci in each driving direction within a traffic light cycle through road monitoring video. The number of vehicles passing through the road stop line when the traffic light is green is counted to obtain the vehicle number. Then, the average time it takes for vehicles to pass the road stop line in each driving direction is obtained by Ti=Tdi / Ci; S3. Obtain the intersection capacity Gi for each driving direction of the designated road using the formula Gi=3600 / Tz*{(Tdi-Ty) / Ti+1}*β, where Ty is a preset fixed value and β is a reduction coefficient, which is 0.

88.

3. The method for predicting bus stop capacity based on video analysis according to claim 2 is characterized in that: All traffic accident videos are queried through a terminal in the office lobby of the local traffic brigade, and the average vehicle speed is the average of the speeds of multiple vehicles detected by the fixed-point test electronic eye.

4. The method for predicting bus stop capacity based on video analysis according to claim 1, characterized in that: Also add the following steps: Step 4: Acquisition of the third data For the shared bicycle parking spaces planned on designated roads, the average difference in the number of shared bicycles before and after rush hour is calculated; Step 5: Comprehensive data acquisition The principles are the same as those of steps 1 to 4. The average value of the platform capacity and number difference of other roads adjacent to the designated road is obtained. Step 6: Comprehensive data analysis The average value of the difference between the number of designated roads and other roads is marked as EPr, and the platform capacity of the designated roads and other roads is GNr, r = 1, 2, ... v, indicating that the total number of roads between the designated roads and other roads is v; Then, the optimization factor for bus stop construction is obtained by Yr={EPr / (EP1+EP2+,...,+EPv)}+GNr. The optimization factor for bus stop construction is the evaluation value of the best location of the bus stop among multiple roads. Step 7: Comprehensive data comparison Then, the preferred factors of the bus stop construction on the designated road are compared with those on other roads, and the road with the largest preferred factor value is obtained as the preferred construction road for the bus stop.

5. The method for predicting bus stop capacity based on video analysis according to claim 4 is characterized by: The specific calculation method for the average value of the difference in the number of shared bicycles in step 4 is: S1. Through questionnaire surveys and interviews, the rush hour is divided into the front and back periods, and the front and back periods are obtained; S2. Collect the number of shared bicycles in the first and second time periods respectively, and then calculate the absolute value of the difference between the number of shared bicycles in the first and second time periods as the number difference. The number of shared bicycles dispatched out and / or transferred in by the shared bicycle dispatcher in the first and second time periods is not included in the calculation; By analogy, in different time periods, multiple groups of quantity differences are obtained and marked as Ek, k = 1, 2, ..., m, indicating that there are m groups of quantity differences; S3. Calculate the average value EP of the quantity difference by EP=(E1+E2+,...,+Em) / m.

Images