A graphic recognition method and recognition system suitable for traffic signs

Abstract

The present invention belongs to the technical field of road traffic, and discloses a graphic recognition method and a recognition system suitable for traffic signs. The method comprises: S1, performing image acquisition on a traffic sign in front of a road where a vehicle is traveling by an image acquisition module to obtain an initial image containing the traffic sign; S2, performing preprocessing on the initial image containing the traffic sign by an image preprocessing module to obtain a graphic outline of the traffic sign; S3, performing recognition processing on the graphic outline of the traffic sign by a graphic recognition module to obtain a classification recognition result of the traffic sign; S4, a message reminder module generates corresponding prompt information according to the recognition result of the traffic sign, and sends it to the vehicle. The present invention automatically recognizes the graphics of the traffic sign to obtain the graphic meaning of the traffic sign, and has the advantages of high efficiency and good accuracy in graphic recognition.

Description

Technical Field

[0001] The present invention belongs to the technical field of road traffic, and in particular relates to a graphic recognition method and a recognition system suitable for traffic signs. Background Art

[0002] Traffic signs are road facilities that use graphics or text to convey guidance, restriction, warning or instruction information. They are an effective measure for traffic management departments to carry out traffic management work, and are also an important method for them to ensure road traffic safety and smoothness. There are many types of traffic signs, including warning signs, prohibition signs, instruction signs, road signs, road construction safety signs, etc. The graphics on different traffic signs are different and represent different meanings. In actual driving, traffic signs cannot actively remind vehicle drivers, and vehicle drivers may neglect traffic signs, and even traffic accidents may occur. Therefore, it is very necessary to study the automatic recognition method of different graphics on traffic signs and provide vehicle drivers with traffic prompt information in a timely manner based on the recognition results. Summary of the invention

[0003] The present invention discloses a graphic recognition method and system suitable for traffic signs. The system actively collects images of traffic signs in front of the road where the vehicle is traveling through an on-board image acquisition module, and sequentially pre-processes and recognizes the collected images containing the traffic signs, thereby ultimately achieving the purpose of automatically recognizing the graphic meaning of the traffic signs. The system can also generate corresponding prompt information and send it to the vehicle to provide help information for the vehicle driver during the driving process.

[0004] To achieve the above-mentioned purpose, the present invention provides the following method steps:

[0005] Step 1: Using an image acquisition module to acquire an image of a traffic sign in front of the road on which the vehicle is traveling, to obtain an initial image containing the traffic sign;

[0006] Step 2: preprocessing the initial image containing the traffic sign through an image preprocessing module to obtain a graphic outline of the traffic sign;

[0007] Step 3: using a graphic recognition module to recognize the graphic outline of the traffic sign to obtain a classification recognition result of the traffic sign;

[0008] Step 4: The message reminder module generates corresponding prompt information according to the recognition result of the traffic sign and sends it to the vehicle;

[0009] In step 3, the graphic outline of the traffic sign is recognized and processed, which specifically includes the following steps:

[0010] The first step is to obtain the approximate polygons of the graphic outlines of different types of traffic signs in turn, and classify the approximate polygons of each type according to the number of sides, so as to divide all the polygons to be identified into different categories of the number of sides;

[0011] The second step is to obtain the initial feature quantity of each polygon of different types and different edge number categories respectively, and cluster the polygons of the same type and the same edge number category based on the initial feature quantity, until all polygons of all types and all edge number categories have been clustered, so as to further divide all polygons to be identified into different initial feature categories;

[0012] Step 3: Determine the number of polygons in each initial feature category. If the number of polygons is 1, the polygons in the initial feature category are no longer classified. If the number of polygons is not 1, continue to perform the following steps;

[0013] Step 4: For polygons of different initial feature categories with different edge numbers under different types, the feature quantities of each polygon are obtained respectively, and polygons of the same initial feature category with the same edge number under the same type are clustered based on the feature quantities, until polygons of all initial feature categories of all edge numbers under all types are clustered, thereby dividing all polygons to be identified into different feature categories;

[0014] Step 5: Determine the number of polygons in each feature category. When the number of polygons is 1, the polygons in the feature category will no longer be classified. When the number of polygons is not 1, if the number of rounds of clustering processing of the polygon feature quantity by the system does not exceed the threshold, continue to execute the previous step, otherwise the step ends.

[0015] Compared with the prior art, the present invention has the following beneficial effects:

[0016] The present invention discloses a method for identifying a graphic of a traffic sign. First, the image of the traffic sign in front of the vehicle is actively collected by an on-board image collection module. Then, the collected image is cropped to obtain an initial image containing the traffic sign. Then, the type of traffic sign to which the initial image belongs is judged based on the main color composition of the initial image, thereby narrowing the recognition range of the graphic of the traffic sign. Then, the initial image is preprocessed and gray-scaled, and an approximate polygon of the graphic outline of the traffic sign is further obtained. The characteristic values ​​of the polygon are extracted hierarchically by calculating the initial characteristic quantity and characteristic quantity of the polygon, and are used to perform multi-level classification of the polygon. The present invention has the advantages of high efficiency and good accuracy in graphic recognition. Finally, the system can automatically obtain the graphic meaning of the traffic sign and provide driving help information for the vehicle driver, thereby solving the problem in the prior art that the traffic sign cannot actively send reminders to the vehicle driver, and the vehicle driver may neglect the traffic sign, and even cause a traffic accident. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a method for pattern recognition of traffic signs according to the present invention;

[0018] Figure 2 A flowchart of the steps of preprocessing an initial image containing a traffic sign in the present invention;

[0019] Figure 3 A flow chart of the steps before the graphic outline of a traffic sign is recognized in the present invention;

[0020] Figure 4 A flowchart of the steps of segmenting the edges of the approximate polygon of the graphic outline of the traffic sign in the present invention;

[0021] Figure 5 A flowchart of the steps of identifying the graphic outline of a traffic sign in the present invention;

[0022] Figure 6 A structural diagram of a graphic recognition system applicable to traffic signs of the present invention;

[0023] Figure 7 It is a composition structure diagram of the image preprocessing module of the present invention;

[0024] Figure 8 It is a composition structure diagram of the pattern recognition module of the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0026] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0027] References Figure 1 As shown, the present invention provides a graphic recognition method applicable to traffic signs, which is implemented by the following steps:

[0028] Step 1: Using an image acquisition module to acquire an image of a traffic sign in front of the road on which the vehicle is traveling, to obtain an initial image containing the traffic sign;

[0029] Step 2: preprocessing the initial image containing the traffic sign through an image preprocessing module to obtain a graphic outline of the traffic sign;

[0030] Step 3: Using a graphic recognition module to recognize the graphic outline of the traffic sign, and obtain a classification recognition result of the traffic sign;

[0031] Step 4: The message reminder module generates corresponding prompt information according to the above-mentioned recognition result of the traffic sign and sends it to the vehicle.

[0032] Furthermore, in the above step one, an image preprocessing module is installed on the vehicle, which is an electronic device with a photo or video recording function, and its shooting angle is adjustable, so as to ensure that the image acquisition module can capture images of traffic signs in front of the vehicle's driving road. Because the captured image screen also includes images other than traffic signs, such as images of vehicles, buildings, etc., the captured image screen also needs to be preliminarily cropped in this step to obtain an initial image containing the traffic sign, so as to facilitate the recognition of the traffic sign graphics in subsequent steps.

[0033] Furthermore, considering that different types of traffic signs often have relatively obvious color differences, for example, warning-type traffic signs generally have a yellow background, black borders, and black patterns, and the most common color is yellow; indicative-type traffic signs generally have a blue background and white patterns, and the most common color is blue. Therefore, based on the main color composition of the traffic sign, the type of traffic sign to which the above-mentioned initial image belongs can be judged. The specific judgment process includes first obtaining the R pixel matrix, G pixel matrix, and B pixel matrix of the initial image containing the traffic sign, and then based on the pixel matrix, counting the most common color values ​​in each pixel of the initial image, and further determining the type of traffic sign to which the initial image belongs based on the color value. Through this step, the initial image can be divided into different traffic sign types, which narrows the recognition range of traffic sign graphics in subsequent steps and improves the system's recognition speed of traffic sign graphics.

[0034] Further, refer to Figure 2 As shown, the image preprocessing module is used to preprocess the initial image containing the traffic sign to obtain the graphic outline of the traffic sign, which specifically includes the following steps:

[0035] The first step is to obtain an R pixel matrix, a G pixel matrix, and a B pixel matrix of an initial image containing a traffic sign, and perform grayscale processing on the initial image based on the pixel matrix to obtain a grayscale image of the initial image;

[0036] Step 2: Find the pixel points whose grayscale values ​​suddenly change from the grayscale image of the initial image, and all the pixel points whose grayscale values ​​suddenly change form the outlines of different figures in the grayscale image respectively;

[0037] The third step is to find a graphic outline that is consistent with a graphic outline preset in the system in the grayscale image containing different graphics, and extract the graphic of the traffic sign contained in the graphic outline, so as to obtain the graphic outline of the traffic sign. The graphic outlines preset in the system include an equilateral triangle, a circle, an octagon, a rectangle, and a square.

[0038] Specifically, in the third step above, considering that traffic signs are generally composed of a border with a relatively obvious shape and special graphics contained in the border to represent special traffic information, the shape of the border is usually an equilateral triangle, circle, octagon, rectangle, or square. Therefore, in this step, in the grayscale image, the graphic contour consistent with these shapes is found, and the graphics contained in the graphic contour are extracted as the graphics of the traffic sign. Finally, the contour of the graphic is extracted as the graphic contour of the traffic sign. In subsequent steps, the graphic contour of the traffic sign is recognized and processed to achieve the purpose of automatically identifying the graphic meaning of the traffic sign.

[0039] Further, refer to Figure 3 As shown, before the graphic outline of the traffic sign is recognized, the following steps are also included:

[0040] The first step is to establish a two-dimensional plane coordinate system. Based on the positional relationship between the pixel points on the graphic contour of the traffic sign in the pixel matrix of the traffic sign graphic, different coordinate points on the coordinate system are generated accordingly, and the convex hull calculation method is used for different coordinate points to obtain an approximate polygon of the graphic contour of the traffic sign.

[0041] The second step is to formally describe the approximate polygon of the graphic outline of the traffic sign, which specifically includes traversing the corners and sides of the polygon in a clockwise direction from any vertex of the polygon, recording the angle value and side length, and the number of sides of the polygon, and obtaining the initial feature value of the approximate polygon;

[0042] The third step is to segment the edges of the approximate polygon of the graphic outline of the traffic sign, specifically including segmenting the edges of the polygon in a clockwise direction starting from the first edge recorded in the initial feature value of the polygon to obtain the feature value of the approximate polygon.

[0043] Specifically, in the second step above, the format of the initial feature quantity of the approximate polygon can be expressed as q = (a1, l1, a2, l2, ..., a n ,l n ), where q is the initial feature value of the approximate polygon, a i , i∈[1,n] is the angle value of different angles of the polygon, l i ,i∈[1,n] is the length of different sides of the polygon, n is the number of sides of the polygon, and the initial feature quantity describes the parameters of different angles and different sides that make up the polygon. Different polygons often have different corresponding initial feature quantities. Therefore, in the subsequent steps, the polygons can be classified according to their different initial feature quantities.

[0044] Further, refer to Figure 4 As shown, in the third step, the edges of the approximate polygon of the graphic outline of the traffic sign are segmented, and the feature quantity of the approximate polygon is obtained, which specifically includes the following steps:

[0045] Step 1: For one edge of the polygon, find the corresponding curve of the figure contour with the same starting point and end point as the edge, and select different points on the curve;

[0046] The second step is to calculate the result value of the characteristic calculation formula according to the different points selected on the curve. The characteristic calculation formula is specifically θ i =(1-(li / L i -1 / 2) 2 )×h i , where θ i is the result value, L i is the side length of the polygon, i∈[1,n], n is the number of sides of the polygon, l i h is the distance from the projection point of the point on the curve to the starting point of the polygon edge, i is the distance from the point on the curve to the edge of the polygon;

[0047] Step 3: Select the point on the curve with the largest corresponding result value and use it as the new vertex of the polygon. Connect the new vertex with the starting point and end point of the original polygon edge. Then one edge of the original polygon is split into two edges, and the l of the original polygon edge is recorded separately. i and h i , and obtain the feature quantity of the approximate polygon.

[0048] Specifically, in the third step above, the format of the feature quantity of the approximate polygon can be expressed as f = (l1, h1, l2, h2, ..., l n ,h n ), where f is the characteristic value of the polygon, l i , i∈[1,n] is the distance from the projection point of the point on the above curve on the polygon edge to the starting point of the polygon edge, h i ,i∈[1,n] is the distance from the point on the above curve to the edge of the polygon, and n is the number of edges of the polygon. Because for some more complex traffic sign graphics, different polygons cannot be distinguished only by the initial feature quantities of their approximate polygons, it is necessary to further extract the features of the polygons by calculating the feature quantities of their approximate polygons, so as to facilitate the continued classification of the polygons.

[0049] Further, refer to Figure 5 As shown, the present invention recognizes and processes the graphic outline of a traffic sign, specifically including the following steps:

[0050] The first step is to obtain the approximate polygons of the graphic outlines of different types of traffic signs in turn, and classify the approximate polygons of each type according to the number of sides, so as to divide all the polygons to be identified into different categories of the number of sides;

[0051] The second step is to obtain the initial feature quantity of each polygon of different types and different edge number categories respectively, and cluster the polygons of the same type and the same edge number category based on the initial feature quantity, until all polygons of all types and all edge number categories have been clustered, so as to further divide all polygons to be identified into different initial feature categories;

[0052] Step 3: Determine the number of polygons in each initial feature category. If the number of polygons is 1, the polygons in the initial feature category are no longer classified. If the number of polygons is not 1, continue to perform the following steps;

[0053] Step 4: For polygons of different initial feature categories with different edge numbers under different types, the feature quantities of each polygon are obtained respectively, and polygons of the same initial feature category with the same edge number under the same type are clustered based on the feature quantities, until polygons of all initial feature categories of all edge numbers under all types are clustered, thereby dividing all polygons to be identified into different feature categories;

[0054] Step 5: Determine the number of polygons in each feature category. When the number of polygons is 1, the polygons in the feature category will no longer be classified. When the number of polygons is not 1, if the number of rounds of clustering processing of the polygon feature quantity by the system does not exceed the threshold, continue to execute the previous step, otherwise the step ends.

[0055] Specifically, in the above steps, clustering processing is performed on polygons according to the initial feature quantity or feature quantity, which can be implemented by a common clustering algorithm in machine learning, such as the K-Means algorithm and the density clustering algorithm. In this embodiment, the specific implementation steps are not repeated. The above steps first classify the approximate polygons of the graphic outline of the traffic sign according to the type of traffic sign to which it belongs, the number of sides of the polygon, and the initial feature quantity of the polygon, so as to divide all the polygons to be identified into different initial feature categories. When the number of polygons in the initial feature category is not equal to 1, the polygons in the initial feature category are continuously classified according to the feature quantity of the polygon, and then all the polygons to be identified are divided into different feature categories, until the number of polygons in the feature category is 1, or the number of rounds of clustering processing of the polygon feature quantity exceeds the threshold, then the system ends the execution of the recognition step;

[0056] After the above-mentioned recognition step is completed, it also includes similarity matching the initial feature quantity or feature quantity of the initial feature category or the polygon in the feature category with the initial feature quantity or feature quantity of the template graphic of the traffic sign pre-stored in the database. The initial feature quantity and feature quantity of the template graphic of the traffic sign are also obtained according to the method for solving the initial feature quantity and feature quantity of the polygon proposed in this embodiment. When the result of the above-mentioned similarity matching is greater than the specified threshold, the system obtains the final recognition result of the traffic sign graphic, and the system automatically obtains the graphic meaning of the traffic sign. At the same time, the system generates corresponding traffic prompt information based on the recognition result, and sends it to the vehicle to provide driving help information for the vehicle driver.

[0057] References Figure 6 As shown, the system also provides a graphic recognition system suitable for traffic signs, which specifically includes the following main modules:

[0058] An image acquisition module is installed on a vehicle to acquire an image of a traffic sign in front of the road on which the vehicle is traveling, and obtain an initial image containing the traffic sign;

[0059] An image preprocessing module, used for preprocessing an initial image containing a traffic sign and obtaining a graphic outline of the traffic sign;

[0060] A graphic recognition module is used to recognize and process the graphic outline of the traffic sign and obtain the classification recognition result of the traffic sign;

[0061] The message reminder module is used to perform similarity matching between the initial feature quantity or feature quantity of the polygon and the feature quantity of the template graphic of the traffic sign pre-stored in the database, obtain the final recognition result of the traffic sign, and generate corresponding prompt information based on the recognition result and send it to the vehicle.

[0062] Further, refer to Figure 7 As shown, the above-mentioned image preprocessing module also includes the following units:

[0063] The first unit is used to obtain a pixel matrix of an initial image containing a traffic sign, and count the most frequently appearing color values ​​in each pixel of the initial image based on the pixel matrix, and determine the type of the traffic sign according to the color value;

[0064] The second unit is used to perform grayscale processing on the initial image, find a graphic outline consistent with a graphic outline preset in the system in the grayscale image, and extract the graphic of the traffic sign contained in the graphic outline, thereby obtaining the graphic outline of the traffic sign.

[0065] Further, refer to Figure 8 As shown, the graphic recognition module includes the following units:

[0066] The third unit is used for using the convex hull calculation method for different points on the graphic contour to obtain an approximate polygon of the graphic contour of the traffic sign, for formally describing the approximate polygon of the graphic contour of the traffic sign to obtain an initial feature quantity of the approximate polygon, and for performing edge segmentation processing on the approximate polygon of the graphic contour of the traffic sign to obtain a feature quantity of the approximate polygon;

[0067] The fourth unit is used to perform a polygon recognition process based on the type, number of sides, initial feature quantity, and feature quantity of the approximate polygon of the graphic outline of the traffic sign, and classify all polygons to be recognized into different categories.

[0068] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0069] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The above-mentioned program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0070] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0071] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0072] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A graphic recognition method applicable to traffic signs, characterized in that: The steps include: S1. Capturing an image of a traffic sign ahead of the road on which the vehicle is traveling by means of an image acquisition module to obtain an initial image containing the traffic sign; S2. Preprocessing the initial image containing the traffic sign by an image preprocessing module to obtain a graphic outline of the traffic sign; S3, using a graphic recognition module to recognize the graphic outline of the traffic sign to obtain a classification recognition result of the traffic sign; S4, the message reminder module generates corresponding prompt information according to the recognition result of the traffic sign, and sends it to the vehicle; Before the image contour of the traffic sign is recognized in S3, the following steps are also included: S31, establishing a two-dimensional plane coordinate system, generating different coordinate points on the coordinate system based on the positional relationship between the pixel points on the graphic contour of the traffic sign in the pixel matrix of the graphic of the traffic sign, and using a convex hull calculation method for the different coordinate points, thereby obtaining an approximate polygon of the graphic contour of the traffic sign; S32. Formalizing the description of the approximate polygon of the graphic outline of the traffic sign, specifically including traversing the corners and sides of the polygon in a clockwise direction from any vertex of the polygon, recording the angle value and the side length, and the number of sides of the polygon, and obtaining the initial feature quantity of the approximate polygon; the format of the initial feature quantity of the approximate polygon is q=(a1,l1,a2,l2,...,a n ,l n ), where q is the initial feature value of the approximate polygon, a i , i∈[1,n] is the angle value of different angles of the polygon, l i ,i∈[1,n] is the length of different sides of the polygon, n is the number of sides of the polygon; S33, performing edge segmentation processing on the approximate polygon of the graphic outline of the traffic sign, specifically comprising starting from the first edge recorded in the initial feature amount of the polygon in a clockwise direction, and segmenting the edges of the polygon respectively to obtain the feature amount of the approximate polygon; In S33, the edges of the approximate polygon of the graphic outline of the traffic sign are segmented to obtain the feature quantity of the approximate polygon, which specifically includes the following steps: S331. For one edge of the polygon, find a corresponding curve of a graphic contour having the same starting point and end point as the edge, and select different points on the curve; S332, according to different points selected on the curve, calculate the result value of the characteristic calculation formula, the characteristic calculation formula is specifically θ i =(1-(l i / L i -1 / 2) 2 )×h i , where θ i is the result value, L i is the side length of the polygon, i∈[1,n], n is the number of sides of the polygon, l i h is the distance from the projection point of the point on the curve to the starting point of the polygon edge, i is the distance from the point on the curve to the edge of the polygon; S333, select the point on the curve with the largest corresponding result value, use it as the new vertex of the polygon, connect the new vertex with the starting point and end point of the original polygon edge, and record the l of the original polygon edge respectively. i and h i , and obtain the feature quantity of the approximate polygon; the format of the feature quantity of the approximate polygon is expressed as f = (l1, h1, l2, h2, ..., l n ,h n ), where f is the characteristic value of the polygon, l i , i∈[1,n] is the distance from the projection point of the point on the above curve on the polygon edge to the starting point of the polygon edge, h i ,i∈[1,n] is the distance from the point on the above curve to the edge of the polygon, and n is the number of edges of the polygon; S3 performs recognition processing on the graphic outline of the traffic sign, which specifically includes the following steps: For the graphic outlines of different types of traffic signs, approximate polygons of each graphic outline are obtained respectively, and the approximate polygons of each type are classified according to the number of sides, so that all polygons to be identified are divided into different categories of the number of sides; For polygons of different types and different edge numbers, the initial feature values ​​of each polygon are obtained respectively, and polygons of the same type and the same edge number category are clustered based on the initial feature values, until polygons of all types and all edge number categories are clustered, so as to further divide all polygons to be identified into different initial feature categories; Determine the number of polygons in each initial feature category. If the number of polygons is 1, the polygons in the initial feature category are no longer classified. If the number of polygons is not 1, continue to perform the following steps; For polygons of different initial feature categories with different numbers of edges under different types, the feature quantities of each polygon are obtained respectively, and polygons of the same initial feature category with the same number of edges under the same type are clustered based on the feature quantities, until polygons of all initial feature categories of all numbers of edges under all types are clustered, thereby dividing all polygons to be identified into different feature categories; Determine the number of polygons in each feature category. When the number of polygons is 1, the polygons in the feature category will no longer be classified. When the number of polygons is not 1, if the number of rounds of clustering processing of the polygon feature quantity by the system does not exceed the threshold, continue to execute the previous step, otherwise the step ends.

2. A method for pattern recognition applicable to traffic signs according to claim 1, characterized in that: Before the initial image containing the traffic sign is preprocessed by the image preprocessing module in S2, it also includes obtaining the R pixel matrix, G pixel matrix, and B pixel matrix of the initial image containing the traffic sign, and based on the pixel matrix, counting the most common color values ​​in each pixel of the initial image, and determining the type of the traffic sign according to the color value.

3. A method for pattern recognition applicable to traffic signs according to claim 2, characterized in that: In S2, the initial image containing the traffic sign is preprocessed by an image preprocessing module, which specifically includes the following steps: S21, obtaining an R pixel matrix, a G pixel matrix, and a B pixel matrix of an initial image containing a traffic sign, and performing grayscale processing on the initial image based on the pixel matrix to obtain a grayscale image of the initial image; S22, finding pixel points whose grayscale values ​​suddenly change from the grayscale image of the initial image, and all pixel points whose grayscale values ​​suddenly change respectively form the outlines of different figures in the grayscale image; S23. In the grayscale image containing different graphics, find a graphic outline that is consistent with a graphic outline preset in the system, and extract the graphic of the traffic sign contained in the graphic outline, so as to obtain the graphic outline of the traffic sign. The graphic outline preset in the system includes an equilateral triangle, a circle, an octagon, a rectangle, and a square.

4. A graphic recognition system for traffic signs, used to implement the method according to claim 3, characterized in that: Includes the following modules: An image acquisition module is installed on a vehicle to acquire an image of a traffic sign in front of the road on which the vehicle is traveling, and obtain an initial image containing the traffic sign; An image preprocessing module, used for preprocessing an initial image containing a traffic sign and obtaining a graphic outline of the traffic sign; A graphic recognition module is used to recognize and process the graphic outline of the traffic sign and obtain the classification recognition result of the traffic sign; The message reminder module is used to perform similarity matching between the initial feature quantity or feature quantity of the polygon and the feature quantity of the template graphic of the traffic sign pre-stored in the database, obtain the final recognition result of the traffic sign, and generate corresponding prompt information based on the recognition result and send it to the vehicle.

5. A graphic recognition system suitable for traffic signs according to claim 4, characterized in that: The image preprocessing module includes the following units: The first unit is used to obtain a pixel matrix of an initial image containing a traffic sign, and count the most frequently appearing color values ​​in each pixel of the initial image based on the pixel matrix, and determine the type of the traffic sign according to the color value; The second unit is used to perform grayscale processing on the initial image, find a graphic outline consistent with a graphic outline preset in the system in the grayscale image, and extract the graphic of the traffic sign contained in the graphic outline, thereby obtaining the graphic outline of the traffic sign.

6. A graphic recognition system suitable for traffic signs according to claim 5, characterized in that: The graphic recognition module includes the following units: The third unit is used for using the convex hull calculation method for different points on the graphic contour to obtain an approximate polygon of the graphic contour of the traffic sign, for formally describing the approximate polygon of the graphic contour of the traffic sign to obtain an initial feature quantity of the approximate polygon, and for performing edge segmentation processing on the approximate polygon of the graphic contour of the traffic sign to obtain a feature quantity of the approximate polygon; The fourth unit is used to perform a polygon recognition process based on the type, number of sides, initial feature quantity, and feature quantity of the approximate polygon of the graphic outline of the traffic sign, and classify all polygons to be recognized into different categories.

Images