High-precision extra-multi-axle large vehicle identification method and system based on discrete square wave segmentation

Abstract

The invention discloses a high-precision extra-multi-axle large vehicle identification method and system based on discrete square wave segmentation, and the method comprises the steps: 1, carrying out the dimension raising preprocessing of the data information of the axle load and axle distance of an extra-multi-axle large vehicle measured by a dynamic weighing system, and forming a two-dimensional square wave discrete diagram; step 2, traversing the discrete points by performing mathematical representation on the position relationship among the square-wave-like discrete points, marking the discrete points with different position relationships by using different symbols, and drawing on a square-wave-like discrete diagram; 3, indexes are added to special points on the square-wave-like discrete diagram, the square-wave-like discrete diagram is divided into a plurality of interval blocks, and the segmentation line between every two interval blocks is the recognition position of the extra-multi-axis large transport vehicle; 4, counting the number of the interval blocks and the number of the discrete points in each interval block, and outputting the number of the interval blocks and the discrete points to calculate the total load of each extra multi-axle large transport vehicle. According to the invention, accurate identification and load calculation of the extra-multi-axle large vehicle are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of large-scale transport vehicle identification, and in particular relates to a high-precision multi-axle large-scale vehicle identification method and system based on discrete quasi-square wave segmentation. Background Art

[0002] With the development of economy and the acceleration of urbanization, large vehicles with multiple axles are increasingly used in logistics, transportation, construction and other fields. However, due to the large size of large vehicles with multiple axles and the complex distribution of axle load and wheelbase, it is difficult to identify the vehicles. At present, there is no mature algorithm in China that can accurately identify multiple large vehicles in a row based on the distribution of axle load and wheelbase of large vehicles when the number of axles is greater than or equal to 6.

[0003] Previous vehicle recognition algorithms were mainly based on image processing technology, which used cameras, scanners and other equipment to collect vehicle images, used image processing algorithms to analyze vehicle images, and extracted feature information such as the vehicle's shape and outline to achieve recognition of large multi-axle vehicles.

[0004] However, this method is often affected by complex environments, such as lighting, weather and other factors that interfere with recognition. If part of the vehicle is camouflaged or blocked by other objects, the difficulty of recognition will increase and the accuracy of recognition will decrease. In addition, the image features of large multi-axle vehicles are relatively complex and difficult to accurately extract and recognize. Therefore, this method is not ideal for the recognition of large multi-axle vehicles, and cannot accurately identify the distribution of vehicle axle loads and wheelbases. Therefore, it is of great significance to propose a high-precision large multi-axle vehicle recognition method and system based on discrete square wave segmentation. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a high-precision identification method and system for large-scale multi-axle vehicles based on discrete quasi-square wave segmentation, which can plot the data on axle loads and wheelbases of multiple large-scale multi-axle vehicles collected by the dynamic weighing system, thereby obtaining a quasi-square wave discrete graph. By performing a series of operations such as marking, indexing, and segmenting the quasi-square wave discrete graph, the purpose of segmenting the quasi-square wave discrete graph into a number of interval blocks is achieved, and then the data information of each large-scale multi-axle vehicle can be accurately segmented from the huge and scattered discrete data, and the total load information of each large-scale multi-axle vehicle can be output accordingly, thereby realizing accurate identification of large-scale multi-axle vehicles.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A high-precision multi-axle heavy vehicle recognition method based on discrete quasi-square wave segmentation comprises the following steps:

[0008] Step 1: Perform dimension-upgrading preprocessing on the data information of the axle load and wheelbase of the multi-axle heavy vehicle measured by the dynamic weighing system to form a two-dimensional square wave discrete graph;

[0009] Step 2: By mathematically representing the positional relationship between the discrete points of the quasi-square wave, traverse the discrete points, mark the discrete points with different positional relationships with different symbols and draw them on the quasi-square wave discrete graph;

[0010] Step 3: Add indexes to the special points on the quasi-square wave discrete graph, divide the quasi-square wave discrete graph into several interval blocks, and the dividing line between two interval blocks is the identification position of the multi-axle heavy-duty transport vehicle;

[0011] Step 4: Count the number of interval blocks and discrete points in each interval block and calculate the total load of each multi-axle heavy-duty transport vehicle.

[0012] Preferably, the step 1 specifically comprises the following steps:

[0013] Step 1.1: The one-dimensional raw data of the axle load and wheelbase of a number of large multi-axle vehicles collected by the dynamic weighing system are processed and converted into two-dimensional data;

[0014] Step 1.2: Based on the two-dimensional data information, the axle load is set as the y-axis, the wheelbase is set as the x-axis, and the two-dimensional graph drawn is a two-dimensional square wave discrete graph.

[0015] Preferably, the step 2 specifically comprises the following steps:

[0016] Step 2.1: Calculate the slope k of the i-th point by using the slope expression i , to express the positional relationship between two adjacent discrete points, along the direction of increasing wheelbase value in the two-dimensional data, with the help of the slope expression, the change of the axle load is expressed one by one from left to right, that is, the slope of the discrete points is traversed and calculated along the x-axis direction in the square wave discrete diagram;

[0017] Step 2.2: According to the different slopes between two adjacent discrete points, use corresponding symbols to mark and plot the discrete points with different positional relationships.

[0018] Preferably, the step 3 specifically comprises the following steps:

[0019] Step 3.1: Put the values ​​of multiple axes on the x-axis in different arrays for the two situations where the difference between two adjacent axle loads distributed along the wheelbase is ≥ 0 and less than -1120 mm and marked with different symbols;

[0020] Step 3.2: Index the x value of one of the special points of the symbol type, and use the values ​​of each two adjacent elements in the array as the initial and end points of the interval block in turn. The initial point, the end point and the middle area are divided into an interval block along the x-axis, and then the entire square wave discrete graph is divided into several interval blocks;

[0021] Step 3.3: Draw a black straight line perpendicular to the x-axis between each two adjacent interval blocks as the dividing line of the interval blocks.

[0022] Preferably, the step 4 specifically comprises the following steps:

[0023] Step 4.1: In the square wave discrete point diagram, the number of large vehicles with multiple axles is directly reflected as the number of interval blocks; the number of axles of each vehicle is directly reflected as the number of discrete points in each interval block;

[0024] Step 4.2: Count the number of segmented and identified vehicles and the number of axles of each large multi-axle vehicle. The specific process is as follows:

[0025] Use the function index to find the jth common element of the two arrays, and use it as the starting point to traverse and compare whether the elements in one array are in the other array, until the (j+1)th common element is encountered, and stop the comparison. Count the number of discrete points in the jth interval block;

[0026] Step 4.3: Output the number of segmented and identified vehicles and the number of axles of each large multi-axle vehicle;

[0027] Step 4.4: Output the sum of the number of all discrete points in each interval block along the y-axis direction, that is, output the total vehicle load of each extra-multi-axle heavy vehicle.

[0028] The present invention also provides a high-precision extra-multi-axle large vehicle identification system based on discrete square wave segmentation, which is used to implement the high-precision extra-multi-axle large vehicle identification method based on discrete square wave segmentation, comprising: a preprocessing module, a drawing module, an identification module and a calculation module;

[0029] The preprocessing module is used to perform dimension preprocessing on the data information of the axle load and wheelbase of the multi-axle heavy vehicle measured by the dynamic weighing system to form a two-dimensional square wave discrete graph;

[0030] The drawing module is used to mathematically represent the positional relationship between the discrete points of the quasi-square wave, traverse the discrete points, mark the discrete points with different positional relationships with different symbols and draw them on the quasi-square wave discrete graph;

[0031] The identification module is used to add indexes to special points on the quasi-square wave discrete graph, and divide the quasi-square wave discrete graph into a number of interval blocks, and the dividing line between two interval blocks is the identification position of the multi-axle heavy-duty transport vehicle;

[0032] The calculation module is used to count the number of interval blocks and discrete points in each interval block and calculate the total load of each multi-axle heavy-duty transport vehicle.

[0033] Preferably, the preprocessing module includes: a dimension upgrading unit and a first drawing unit;

[0034] The dimension upgrading unit is used to convert the one-dimensional raw data information of the axle load and wheelbase of a plurality of continuous multi-axle heavy vehicles measured and collected by the dynamic weighing system into two-dimensional data information after processing;

[0035] The first drawing unit is used to set the axle load as the y-axis and the wheelbase as the x-axis based on the two-dimensional data information, and the drawn two-dimensional graph is a two-dimensional square wave discrete graph.

[0036] Preferably, the drawing module includes: a calculation unit and a marking unit;

[0037] The calculation unit is used to calculate the slope k of the i-th point by using the slope expression i , to express the positional relationship between two adjacent discrete points, along the direction of increasing wheelbase value in the two-dimensional data, with the help of the slope expression, the change of the axle load is expressed one by one from left to right, that is, the slope of the discrete points is traversed and calculated along the x-axis direction in the square wave discrete diagram;

[0038] The marking unit is used to mark and draw discrete points with different position relationships with corresponding symbols according to different slope conditions between two adjacent discrete points.

[0039] Preferably, the recognition module comprises: a classification unit, a segmentation unit and a second drawing unit;

[0040] The classification unit is used to place the values ​​of the multiple axes on the x-axis in different arrays in two situations where the difference between two adjacent axle loads distributed along the wheelbase and marked with different symbols is ≥0 and less than -1120 mm;

[0041] The segmentation unit is used to index the x value of one of the special points of the symbol type, and use the values ​​of each two adjacent elements in the array as the initial and end points of the interval block in turn, and segment the initial point, the end point and the middle area into an interval block along the x-axis, thereby segmenting the entire quasi-square wave discrete graph into a plurality of interval blocks;

[0042] The second drawing unit is used to draw a black straight line perpendicular to the x-coordinate axis between every two adjacent interval blocks as a dividing line of the interval blocks.

[0043] Preferably, the calculation module includes: a setting unit, a statistical unit, a first output unit and a second output unit;

[0044] The setting unit is used to intuitively reflect the number of large multi-axle vehicles as the number of interval blocks in the quasi-square wave discrete point diagram; the number of axles of each vehicle is intuitively reflected as the number of discrete points in each interval block;

[0045] The statistical unit is used to count the number of segmented and identified vehicles and the number of axles of each large multi-axle vehicle; the specific process is:

[0046] Use the function index to find the jth common element of the two arrays, and use it as the starting point to traverse and compare whether the elements in one array are in the other array, until the (j+1)th common element is encountered, and stop the comparison. Count the number of discrete points in the jth interval block;

[0047] The first output unit is used to output the number of segmented and identified vehicles counted, and the number of axles of each large multi-axle vehicle;

[0048] The second output unit is used to output the sum of the quantities of all discrete points in each interval block along the y-axis direction, that is, to output the total vehicle load of each extra-multi-axle heavy vehicle.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] (1) The present invention can improve the recognition accuracy of large multi-axle vehicles and make the dynamic weighing system more intelligent, and provides an effective method for representing the position relationship between two discrete points and segmenting square wave discrete graphs in the field of one-dimensional signal processing;

[0051] (2) The present invention realizes the identification of multiple extra-multi-axle large vehicles by segmenting the two-dimensional square wave discrete graph composed of the discrete data of the axle load and wheelbase of multiple extra-multi-axle large vehicles. It solves the problem of identifying large vehicles with ≥ 6 axles, and makes the identification accuracy rate reach more than 95% and the algorithm response time within 1s, providing important technical support for logistics warehousing, transportation and other industries;

[0052] (3) The present invention is not affected by complex external environments, can identify different types of vehicles and calculate vehicle loads, and does not require replacement of hardware equipment, which can greatly reduce economic costs;

[0053] (4) The algorithm designed in the present invention can be improved and optimized according to actual needs to adapt to different application scenarios and has good scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0055] Figure 1 This is a flow chart of a high-precision multi-axle heavy vehicle recognition method based on discrete quasi-square wave segmentation according to an embodiment of the present invention;

[0056] Figure 2 This is a structural diagram of the actual recognition process of a high-precision multi-axle large vehicle recognition method based on discrete square-wave segmentation in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] 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.

[0058] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] Embodiment 1

[0060] like Figure 1 As shown, the present invention provides a high-precision multi-axle large vehicle recognition method based on discrete square wave segmentation, comprising the following steps:

[0061] Step 1: Perform dimension-upgrading preprocessing on the data information of the axle load and wheelbase of the multi-axle heavy vehicle measured by the dynamic weighing system to form a two-dimensional square wave discrete graph;

[0062] Step 2: By mathematically representing the positional relationship between the discrete points of the quasi-square wave, traverse the discrete points, mark the discrete points with different positional relationships with different symbols and draw them on the quasi-square wave discrete graph;

[0063] Step 3: Add indexes to the special points on the quasi-square wave discrete graph, divide the quasi-square wave discrete graph into several interval blocks, and the dividing line between two interval blocks is the identification position of the multi-axle heavy-duty transport vehicle;

[0064] Step 4: Count the number of interval blocks and discrete points in each interval block and calculate the total load of each multi-axle heavy-duty transport vehicle.

[0065] In this embodiment, step 1 specifically includes the following steps:

[0066] Step 1.1: Perform dimension-upgrading preprocessing on the data information of the axle load and wheelbase of the multi-axle heavy vehicle measured by the dynamic weighing system;

[0067] Step 1.2: Draw a two-dimensional square wave-like discrete graph in a coordinate system using the data after the dimensionality upgrade process in step 1.1.

[0068] As a preferred technical solution, the "extremely multi-axle" of the particularly multi-axle large vehicle in step 1.1 specifically refers to a description of a large vehicle with axles far exceeding the normal situation, which is usually not less than 6 axles, and commonly has more than ten axles, more than twenty axles, and may even reach 40 axles or even more.

[0069] As a preferred technical solution, the data measured by the dynamic weighing system is pre-processed by dimensional upgrading in step 1.1, specifically, the one-dimensional raw data information of the axle load and wheelbase of a plurality of continuous multi-axle large vehicles collected by the dynamic weighing system after measurement is processed, sorted, etc., and then converted into two-dimensional data information;

[0070] As a preferred technical solution, the one-dimensional original data of axle load and wheelbase refers specifically to one-dimensional data directly obtained from a dynamic weighing system, in which data of multiple vehicles overlap with each other and the distribution of axle load and wheelbase is in a disordered state;

[0071] As a preferred technical solution, the two-dimensional data information specifically refers to that after the dimension-upgrading process, the wheelbases are arranged in a regular pattern of increasing from left to right, and the axle loads are distributed corresponding to the values ​​of the wheelbases based on the one-dimensional original data, thereby forming two rows of two-dimensional data of the same dimension, in which the axle loads and the wheelbases are distributed correspondingly;

[0072] As a preferred technical solution, the one-dimensional raw data information of the axle load and wheelbase of multiple large multi-axle vehicles collected by the dynamic weighing system is processed and sorted, specifically including the following steps:

[0073] (1) Following the principle of time sequence, multiple overlapping data collected by the dynamic weighing system are separated, and then the axle load and wheelbase data of each multi-axle heavy vehicle are obtained;

[0074] (2) The data of each large multi-axle vehicle is processed, and the original mixed and disordered data of axle load and wheelbase are split into separate distribution forms of axle load and wheelbase. During the splitting process, it is imperative to ensure that each wheelbase corresponds to the axle load at the corresponding position, and the original position of the wheelbase or axle load must not be swapped;

[0075] (3) Arrange the wheelbases of each vehicle in rows in chronological order, and add a random number ranging from 1000 to 6000 mm at the end of each row as the distance between each two adjacent multi-axle heavy vehicles;

[0076] (4) Considering that a wheelbase is generated between every two axle loads, in order to avoid the difference in the dimensions of the axle load data and the wheelbase data after the dimensionality increase processing, the front end of the first row of wheelbase data is assigned a value of 0, indicating that the distribution of the wheelbase data starts at 0;

[0077] (5) In order from left to right and from top to bottom, the wheelbase of each large vehicle with multiple axles is accumulated and processed so that they are arranged in rows, so as to facilitate the parallel distribution of the wheelbases of all large vehicles with multiple axles and change their original aliasing distribution form. The specific operation is: the first wheelbase value is its own value of 0, and starting from the second wheelbase value, the current wheelbase is added to the adjacent previous wheelbase value in sequence as the adjacent next wheelbase value. For two adjacent vehicles, the distance between the two vehicles is added to the wheelbase value at the end of the front vehicle as the wheelbase value at the beginning of the rear vehicle, thereby completing the connection of the wheelbase accumulation process of the two vehicles, and finally making the wheelbase data gradually increase until the accumulation of the wheelbase data of all large vehicles with multiple axles is completed;

[0078] (6) Finally, the axle load data of each vehicle is arranged in rows and distributed corresponding to the wheelbase data, ultimately forming two rows of two-dimensional data with the same dimension and corresponding distribution of axle load and wheelbase.

[0079] As a preferred technical solution, the square wave-like discrete graph drawn in the step 1.2 specifically refers to a two-dimensional graph drawn by setting the axle load as the ordinate (y-axis) and the wheelbase as the abscissa (x-axis) based on the two-dimensional data after dimensionality upgrade processing. Each discrete point in the graph represents a certain axle of a certain multi-axle large vehicle, and the coordinate position information of the discrete point in the graph represents the axle load of a certain axle distributed with the wheelbase. By connecting all the discrete points, a graph similar to a square wave shape can be drawn, so it is called a square wave-like discrete graph. The shape of the graph reflects the distribution of the axle load of a multi-axle large vehicle with the change of wheelbase. The physical meaning of the entire square wave-like discrete graph is to separate the originally mixed and disordered data of multiple multi-axle large vehicles, so that the data of each large vehicle can be distributed in parallel and presented in the same graph.

[0080] In this embodiment, step 2 specifically includes the following steps:

[0081] Step 2.1: Use mathematical expressions to represent the positional relationship between discrete points of the quasi-square wave, and traverse and calculate the discrete points along the x-axis direction;

[0082] Step 2.2: Mark the discrete points with different positional relationships with different symbols and plot them on the quasi-square wave discrete graph.

[0083] As a preferred technical solution, the positional relationship between discrete points in step 2.1 specifically refers to the positional distribution relationship between two adjacent discrete points in the square wave-like discrete graph. The coordinates of the discrete points in the square wave-like discrete graph represent the axle load of a certain axle of the vehicle distributed along the wheelbase. In order to represent the magnitude relationship of the axle load of two adjacent axles of the vehicle as the wheelbase changes, it is intuitively converted into a representation of the positional relationship between the square wave-like discrete points, that is, specifically, the (i+1)th point is located above or below the i-th point, or is exactly parallel to the i-th point;

[0084] As a preferred technical solution, the positional relationship between the discrete points in step 2.1 is mathematically represented, specifically, by calculating the slope k of the i-th point using the slope expression i , to express the positional relationship between two adjacent discrete points. The slope expression is

[0085]

[0086] Where: x i With y i 、x i+1 With y i+1 are the coordinate values ​​of the i-th and i+1-th discrete points respectively.

[0087] As a preferred technical solution, traversing and calculating discrete points along the x-axis in the step 2.1 actually means that along the direction of increasing wheelbase values ​​in the two-dimensional data, the change in the axle load is represented one by one from left to right with the help of the slope expression. Here, the problem is intuitively converted into traversing and calculating the slope of discrete points along the x-axis in a square wave-like discrete graph.

[0088] As a preferred technical solution, in the step 2.2, different symbols are respectively marked for discrete points of different positional relationships, and they are plotted on a quasi-square wave discrete diagram. Specifically, with the distribution of wheelbase, there are different situations for the difference in axle loads between two adjacent axles, that is, the value of the axle load of the next adjacent axle minus the axle load of the current axle is greater than or equal to 0 or less than a certain value. In order to distinguish between these two types of situations, the problem is intuitively converted into marking different symbols for discrete points of different positional relationships and plotting them. When marking discrete points on a quasi-square wave discrete diagram, discrete points of different positional relationships are marked and plotted with corresponding symbols based on the different slopes between two adjacent discrete points;

[0089] As a preferred technical solution, the different symbols specifically refer to the two different symbols, "o" and "*", used to mark the discrete points in the present invention. The marking rule is: when the slope calculation result is less than 0, and the difference between the y value of the (i+1)th point and the i-th point is less than -1120 mm, the discrete point is marked with an "o" symbol; the calculation results of other cases are marked with an "*" symbol.

[0090] In this embodiment, step 3 specifically includes the following steps:

[0091] Step 3.1: Store the two discrete points of different symbol types in two different arrays;

[0092] Step 3.2: Divide the entire quasi-square wave discrete graph into several interval blocks by adding an index to the x value of a special point of one of the symbol types;

[0093] Step 3.3: Draw a black straight line perpendicular to the x-axis between each two adjacent interval blocks as the dividing line of the interval blocks.

[0094] As a preferred technical solution, the two discrete points of different sign types in step 3.1 actually refer to multiple axles in the two cases where the difference between two adjacent axle loads distributed along the wheelbase is ≥ 0 and is less than -1120 mm;

[0095] As a preferred technical solution, in step 3.1, the discrete points are stored in two different arrays. Specifically, in order to visually identify and distinguish each large multi-axle vehicle on the quasi-square wave discrete graph, the values ​​of the two types of points marked with different symbols on the x-axis need to be placed in different arrays, respectively, to facilitate subsequent index segmentation;

[0096] As a preferred technical solution, the two different arrays have the characteristics of different array sizes but the same elements.

[0097] As a preferred technical solution, the process of dividing the quasi-square wave discrete graph into several interval blocks in step 3.2 specifically includes indexing the x value of one of the special points of the symbol type, taking the values ​​of each two adjacent elements in the array as the initial and end points of the interval block in turn, and dividing the initial point, the end point and the middle area into an interval block along the x-axis. The same method is used to divide the entire quasi-square wave discrete graph into several interval blocks;

[0098] As a preferred technical solution, one of the symbol type special points mentioned above actually refers to discrete points marked with an "o" symbol. Specifically, if the calculated slope of the i-th point is less than 0, and the difference in y value between the (i+1)th point and the i-th point is less than -1120 mm, then the (i+1)th point is a special point;

[0099] As a preferred technical solution, the interval blocks actually represent large vehicles with many axles, and dividing the entire square-wave discrete graph into several interval blocks is to intuitively separate each large vehicle with many axles from the square-wave discrete graph.

[0100] As a preferred technical solution, the segmentation line of the interval block in step 3.3 is actually the position where the data information of multiple continuous multi-axle heavy-duty transport vehicles is segmented and identified. In order to intuitively represent the segmentation and identification results of the multi-axle heavy-duty vehicles, a black straight line is drawn between two adjacent vehicles as the vehicle segmentation line.

[0101] In this embodiment, step 4 specifically includes the following steps:

[0102] Step 4.1: Count the number of interval blocks and discrete points in each interval block;

[0103] Step 4.2: Output the interval block and the number of discrete points in each interval block;

[0104] Step 4.3: Output the sum of the number of all discrete points along the y-axis in each interval block.

[0105] As a preferred technical solution, the counting of the number of interval blocks and the number of discrete points in each interval block in step 4.1 specifically means that in the square wave-like discrete point diagram, the number of large vehicles with multiple axles is intuitively reflected as the number of interval blocks; the number of axles of each vehicle is intuitively reflected as the number of discrete points in each interval block. Therefore, counting the number of interval blocks and the number of discrete points in each interval block is actually counting the number of segmented and identified vehicles and the number of axles of each large vehicle with multiple axles;

[0106] As a preferred technical solution, the process of counting the number of discrete points in each interval block specifically refers to using the function index to find the jth identical element of the two arrays in step 3.1, and using it as the starting point, traversing and comparing whether the elements in one array are in the other array in sequence, until the (j+1)th identical element is encountered and the comparison is stopped. In this way, the number of discrete points in the jth interval block is counted.

[0107] As a preferred technical solution, the output of the interval blocks and the number of discrete points in each interval block in the step 4.2 specifically refers to the output of the number of segmented and identified vehicles counted, as well as the number of axles of each multi-axle large vehicle.

[0108] As a preferred technical solution, the sum of the number of all discrete points in each interval block along the y-axis direction is output in step 4.3, specifically, in the square wave discrete point diagram, the total load of each vehicle is intuitively reflected by the sum of the number of all discrete points in each interval block along the y-axis direction. Therefore, outputting the sum of the number of all discrete points in each interval block along the y-axis direction is actually outputting the total vehicle load of each multi-axle heavy vehicle;

[0109] As a preferred technical solution, the sum of the quantities along the y-axis direction actually refers to the total load of all axles of each extra-multi-axle heavy-duty vehicle.

[0110] Embodiment 2

[0111] This embodiment takes the identification of a multi-axle heavy-duty transport vehicle on a highway as an example. Figure 2 The actual recognition process structure diagram of a high-precision multi-axis large vehicle recognition algorithm based on discrete square wave segmentation includes a highway with pre-embedded dynamic weighing system sensors, a server with a data processing system and a data segmentation calculation system, a display, etc.

[0112] Among them, the dynamic weighing system sensors embedded in the highway can collect and transmit the axle load and wheelbase information of the multi-axle heavy-duty transport vehicles. The data processing system of the server mainly processes and organizes the data collected by the sensor, completes the dimension-upgrading operation, and obtains data that can be directly recognized, imported and extracted by the data segmentation and calculation system. The data segmentation and calculation system of the server is a system composed of the multi-axle heavy-duty vehicle identification algorithm designed by the present invention as the main part. Its purpose is to separate the data information about the axle load and wheelbase belonging to the same vehicle from the huge and scattered discrete data, calculate the total load of the vehicle and output the axle load and wheelbase information of this vehicle. The display draws the segmented square wave discrete graph and displays the output results of the data segmentation and calculation system in real time.

[0113] In the actual identification process, when a multi-axle large-scale transport vehicle passes through a highway section with pre-buried dynamic weighing system sensors, the sensor belt will collect relevant data information on the vehicle's axle load and wheelbase, and then transmit the collected data information to the server's data processing system. The processed and sorted two-dimensional data is then imported into the data segmentation calculation system for segmentation calculation. Finally, the segmented square wave discrete graph and the vehicle axle load and wheelbase information are summarized and output through the display.

[0114] Embodiment 3

[0115] This embodiment takes the representation of the positional relationship of discrete points or the segmentation of a two-dimensional square wave-like discrete graph as an example. In the present invention, the positional relationship between two discrete points is represented by calculating the slope of the two discrete points. Then the calculation is performed traversally, and different symbols are marked on the discrete points in turn according to different slope conditions. The discrete points with different symbols are then indexed, and the entire square wave-like discrete graph is segmented into several interval blocks through the index, and the number of discrete points in each interval block is counted. This idea of ​​representing the positional relationship between two discrete points or segmenting a two-dimensional square wave-like discrete graph is very important.

[0116] Embodiment 4

[0117] The present invention also provides a high-precision extra-multi-axle large vehicle identification system based on discrete-like square wave segmentation, which is used to implement the high-precision extra-multi-axle large vehicle identification method based on discrete-like square wave segmentation described in Example 1, including: a preprocessing module, a drawing module, a recognition module and a calculation module;

[0118] The preprocessing module is used to perform dimension preprocessing on the data information of the axle load and wheelbase of the multi-axle heavy vehicle measured by the dynamic weighing system to form a two-dimensional square wave discrete graph;

[0119] The drawing module is used to mathematically represent the positional relationship between the discrete points of the quasi-square wave, traverse the discrete points, mark the discrete points with different positional relationships with different symbols and draw them on the quasi-square wave discrete graph;

[0120] The recognition module is used to add indexes to special points on the quasi-square wave discrete graph, and divide the quasi-square wave discrete graph into several interval blocks. The dividing line between two interval blocks is the identification position of the multi-axle heavy-duty transport vehicle.

[0121] The calculation module is used to count the number of interval blocks and discrete points in each interval block and calculate the total load of each multi-axle heavy-duty transport vehicle.

[0122] In this embodiment, the preprocessing module includes: a dimension upgrading unit and a first drawing unit;

[0123] The dimension upgrading unit is used to convert the one-dimensional raw data information of the axle load and wheelbase of multiple large multi-axle vehicles collected by the dynamic weighing system into two-dimensional data information after processing;

[0124] The first drawing unit is used to set the axle load as the y-axis and the wheelbase as the x-axis based on the two-dimensional data information, and the drawn two-dimensional graph is a two-dimensional square wave discrete graph.

[0125] In this embodiment, the drawing module includes: a calculation unit and a marking unit;

[0126] The calculation unit is used to calculate the slope k of the i-th point by using the slope expression i , to express the positional relationship between two adjacent discrete points, along the direction of increasing wheelbase value in the two-dimensional data, with the help of the slope expression, the change of the axle load is expressed one by one from left to right, that is, the slope of the discrete points is traversed and calculated along the x-axis direction in the square wave discrete diagram;

[0127] The marking unit is used to mark and draw discrete points with different position relationships with corresponding symbols according to different slopes between two adjacent discrete points.

[0128] In this embodiment, the recognition module includes: a classification unit, a segmentation unit and a second drawing unit;

[0129] The classification unit is used to place the values ​​of multiple axes on the x-axis in different arrays in the two cases where the difference between two adjacent axle loads distributed along the wheelbase and marked with different symbols is ≥ 0 and less than -1120 mm;

[0130] The segmentation unit is used to index the x value of one of the special points of the symbol type, and use the values ​​of each two adjacent elements in the array as the initial and end points of the interval block in turn, and segment the initial point, the end point and the middle area into an interval block along the x-axis, thereby segmenting the entire quasi-square wave discrete graph into several interval blocks;

[0131] The second drawing unit is used to draw a black straight line perpendicular to the x-coordinate axis between every two adjacent interval blocks as a dividing line of the interval blocks.

[0132] In this embodiment, the calculation module includes: a setting unit, a statistical unit, a first output unit and a second output unit;

[0133] The setting unit is used to intuitively reflect the number of large vehicles with multiple axles as the number of interval blocks in the quasi-square wave discrete point diagram; the number of axles of each vehicle is intuitively reflected as the number of discrete points in each interval block;

[0134] The statistical unit is used to count the number of segmented and identified vehicles and the number of axles of each large multi-axle vehicle; the specific process is:

[0135] Use the function index to find the jth common element of the two arrays, and use it as the starting point to traverse and compare whether the elements in one array are in the other array, until the (j+1)th common element is encountered, and stop the comparison. Count the number of discrete points in the jth interval block;

[0136] The first output unit is used to output the number of segmented and identified vehicles counted, and the number of axles of each large multi-axle vehicle;

[0137] The second output unit is used to output the sum of the quantities of all discrete points in each interval block along the y-axis direction, that is, to output the total vehicle load of each extra-multi-axle heavy vehicle.

[0138] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A high-precision multi-axle heavy vehicle recognition method based on discrete square wave segmentation. It is characterized in that The following steps are involved: Step 1: Perform dimension-upgrading preprocessing on the data information of the axle load and wheelbase of the multi-axle heavy vehicle measured by the dynamic weighing system to form a two-dimensional square wave discrete graph; Step 2: By mathematically representing the positional relationship between the discrete points of the quasi-square wave, traverse the discrete points, mark the discrete points with different positional relationships with different symbols and draw them on the quasi-square wave discrete graph; Step 3: Add indexes to the special points on the quasi-square wave discrete graph, divide the quasi-square wave discrete graph into several interval blocks, and the dividing line between two interval blocks is the identification position of the multi-axle heavy-duty transport vehicle; Step 4: Count the number of interval blocks and discrete points in each interval block and calculate the total load of each multi-axle heavy-duty transport vehicle.

2. According to the high-precision multi-axle large vehicle recognition method based on discrete quasi-square wave segmentation of claim 1, It is characterized in that The step 1 specifically comprises the following steps: Step 1.1: The one-dimensional raw data of the axle load and wheelbase of a number of large multi-axle vehicles collected by the dynamic weighing system are processed and converted into two-dimensional data; Step 1.2: Based on the two-dimensional data information, the axle load is set as the y-axis, the wheelbase is set as the x-axis, and the two-dimensional graph drawn is a two-dimensional square wave discrete graph.

3. The high-precision multi-axle large vehicle recognition method based on discrete quasi-square wave segmentation according to claim 1, It is characterized in that The step 2 specifically includes the following steps: Step 2.1: Calculate the slope k of the i-th point by using the slope expression i , to express the positional relationship between two adjacent discrete points, along the direction of increasing wheelbase value in the two-dimensional data, with the help of the slope expression, the change of the axle load is expressed one by one from left to right, that is, the slope of the discrete points is traversed and calculated along the x-axis direction in the square wave discrete diagram; Step 2.2: According to the different slopes between two adjacent discrete points, use corresponding symbols to mark and plot the discrete points with different positional relationships.

4. The high-precision multi-axle heavy vehicle recognition method based on discrete quasi-square wave segmentation according to claim 1, It is characterized in that The step 3 specifically comprises the following steps: Step 3.1: Put the values ​​of multiple axes on the x-axis in different arrays for the two situations where the difference between two adjacent axle loads distributed along the wheelbase is ≥ 0 and less than -1120 mm and marked with different symbols; Step 3.2: Index the x value of one of the special points of the symbol type, and use the values ​​of each two adjacent elements in the array as the initial and end points of the interval block in turn. The initial point, the end point and the middle area are divided into an interval block along the x-axis, and then the entire square wave discrete graph is divided into several interval blocks; Step 3.3: Draw a black straight line perpendicular to the x-axis between each two adjacent interval blocks as the dividing line of the interval blocks.

5. The high-precision multi-axle heavy vehicle recognition method based on discrete quasi-square wave segmentation according to claim 4, It is characterized in that The step 4 specifically comprises the following steps: Step 4.1: In the square wave discrete point diagram, the number of large vehicles with multiple axles is directly reflected as the number of interval blocks; the number of axles of each vehicle is directly reflected as the number of discrete points in each interval block; Step 4.2: Count the number of segmented and identified vehicles and the number of axles of each large multi-axle vehicle. The specific process is as follows: Use the function index to find the jth common element of the two arrays, and use it as the starting point to traverse and compare whether the elements in one array are in the other array, until the (j+1)th common element is encountered, and stop the comparison. Count the number of discrete points in the jth interval block; Step 4.3: Output the number of segmented and identified vehicles and the number of axles of each large multi-axle vehicle; Step 4.4: Output the sum of the number of all discrete points in each interval block along the y-axis direction, that is, output the total vehicle load of each extra-multi-axle heavy vehicle.

6. A high-precision multi-axle large vehicle recognition system based on discrete square wave segmentation, It is characterized in that A high-precision multi-axle heavy vehicle recognition method based on discrete square wave segmentation for implementing any one of claims 1 to 5, comprising: a preprocessing module, a drawing module, a recognition module and a calculation module; The preprocessing module is used to perform dimension preprocessing on the data information of the axle load and wheelbase of the multi-axle heavy vehicle measured by the dynamic weighing system to form a two-dimensional square wave discrete graph; The drawing module is used to mathematically represent the positional relationship between the discrete points of the quasi-square wave, traverse the discrete points, mark the discrete points with different positional relationships with different symbols and draw them on the quasi-square wave discrete graph; The identification module is used to add indexes to special points on the quasi-square wave discrete graph, and divide the quasi-square wave discrete graph into a number of interval blocks, and the dividing line between two interval blocks is the identification position of the multi-axle heavy-duty transport vehicle; The calculation module is used to count the number of interval blocks and discrete points in each interval block and calculate the total load of each multi-axle heavy-duty transport vehicle.

7. The high-precision multi-axle large vehicle recognition system based on discrete quasi-square wave segmentation according to claim 6, It is characterized in that The preprocessing module includes: a dimension upgrading unit and a first drawing unit; The dimension upgrading unit is used to convert the one-dimensional raw data information of the axle load and wheelbase of a plurality of continuous multi-axle heavy vehicles measured and collected by the dynamic weighing system into two-dimensional data information after processing; The first drawing unit is used to set the axle load as the y-axis and the wheelbase as the x-axis based on the two-dimensional data information, and the drawn two-dimensional graph is a two-dimensional square wave discrete graph.

8. The high-precision multi-axle large vehicle recognition system based on discrete quasi-square wave segmentation according to claim 6, It is characterized in that The drawing module includes: a calculation unit and a marking unit; The calculation unit is used to calculate the slope k of the i-th point by using the slope expression i , to express the positional relationship between two adjacent discrete points, along the direction of increasing wheelbase value in the two-dimensional data, with the help of the slope expression, the change of the axle load is expressed one by one from left to right, that is, the slope of the discrete points is traversed and calculated along the x-axis direction in the square wave discrete diagram; The marking unit is used to mark and draw discrete points with different position relationships with corresponding symbols according to different slope conditions between two adjacent discrete points.

9. The high-precision multi-axle large vehicle recognition system based on discrete quasi-square wave segmentation according to claim 6, It is characterized in that The recognition module includes: a classification unit, a segmentation unit and a second drawing unit; The classification unit is used to place the values ​​of the multiple axes on the x-axis in different arrays in two situations where the difference between two adjacent axle loads distributed along the wheelbase and marked with different symbols is ≥0 and less than -1120 mm; The segmentation unit is used to index the x value of one of the special points of the symbol type, and use the values ​​of each two adjacent elements in the array as the initial and end points of the interval block in turn, and segment the initial point, the end point and the middle area into an interval block along the x-axis, thereby segmenting the entire quasi-square wave discrete graph into a plurality of interval blocks; The second drawing unit is used to draw a black straight line perpendicular to the x-coordinate axis between every two adjacent interval blocks as a dividing line of the interval blocks.

10. The high-precision multi-axle large vehicle recognition system based on discrete quasi-square wave segmentation according to claim 9, It is characterized in that The calculation module includes: a setting unit, a statistical unit, a first output unit and a second output unit; The setting unit is used to intuitively reflect the number of large multi-axle vehicles as the number of interval blocks in the quasi-square wave discrete point diagram; the number of axles of each vehicle is intuitively reflected as the number of discrete points in each interval block; The statistical unit is used to count the number of segmented and identified vehicles and the number of axles of each large multi-axle vehicle; the specific process is: Use the function index to find the jth common element of the two arrays, and use it as the starting point to traverse and compare whether the elements in one array are in the other array, until the (j+1)th common element is encountered, and stop the comparison. Count the number of discrete points in the jth interval block; The first output unit is used to output the number of segmented and identified vehicles counted, and the number of axles of each large multi-axle vehicle; The second output unit is used to output the sum of the quantities of all discrete points in each interval block along the y-axis direction, that is, to output the total vehicle load of each extra-multi-axle heavy vehicle.

Images