Road surface guiding arrow identification method, device, system and readable storage medium

Abstract

The application discloses a road guiding arrow recognition method, device and system and a readable storage medium. The method comprises the following steps: photographing a road surface to obtain an arrow tip and an arrow bottom of the road surface; constructing a current arrow according to the arrow tip and the arrow bottom; and if it is determined that the reliability of the current arrow meets a preset condition, regarding the current arrow as a road guiding arrow. According to the application, the arrow tip and the arrow bottom of the road surface are obtained to construct the current arrow, and the current arrow is determined as the road guiding arrow when it is determined that the reliability of the current arrow meets the preset condition, so that the road guiding arrow can be recognized even if the road guiding arrow is blocked and cannot be observed completely, and the accuracy of the road guiding arrow recognition is improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, and in particular to a method, device, system, and readable storage medium for recognizing road directional arrows. Background Technology

[0002] Existing road directional arrow recognition typically relies on methods such as IPM (Inverse Perspective Mapping) or LiDAR to obtain the road surface BEV (Bird's Eye View), and then uses deep learning or template matching to output the attributes and positions of single or compound arrows from the bird's eye view.

[0003] However, in congested scenarios with heavy traffic, the road directional arrows are easily obscured, making them impossible to observe completely. In such cases, identifying road directional arrows based on road surface BEVs can lead to lower accuracy in road directional arrow recognition.

[0004] Therefore, improving the accuracy of road directional arrow recognition is an urgent problem to be solved. Summary of the Invention

[0005] The main objective of this invention is to provide a method, device, system, and readable storage medium for recognizing road directional arrows, aiming to improve the accuracy of road directional arrow recognition.

[0006] To achieve the above objectives, the present invention provides a method for recognizing road directional arrows, the method comprising the following steps:

[0007] Photograph the road surface to capture the tip and base of the arrows on the road surface;

[0008] The current arrow is constructed based on the arrow tip and the arrow base. If the credibility of the current arrow meets the preset conditions, the current arrow is used as the road guide arrow.

[0009] Optionally, the step of constructing the current arrow based on the arrow tip and the arrow base includes:

[0010] Obtain the lane lines on the road surface, and determine the first position information of the arrow tip and the second position information of the arrow base;

[0011] Based on the lane lines, the first location information, and the second location information, the arrow tip and the arrow base are grouped;

[0012] Construct the current arrow based on the grouping results.

[0013] Optionally, the step of grouping the arrow tip and the arrow base according to the lane line, the first position information, and the second position information includes:

[0014] Based on the lane lines, the first position information, and the second position information, determine whether the relative position of the arrow tip and the arrow base conforms to a preset rule;

[0015] Based on the first location information and the second location information, determine the line connecting the tip and the base of the arrow, and determine whether the line intersects with the lane line;

[0016] Based on the first location information and the second location information, determine whether the distance between the arrow tip and the arrow base is less than a preset distance threshold;

[0017] If it is determined that the relative position conforms to the preset rules, the connecting line does not intersect the lane line, and the distance is less than the preset distance threshold, then the arrow tip and the arrow base are grouped together.

[0018] Optionally, after the step of constructing the current arrow based on the arrow tip and the arrow base, the method includes:

[0019] Detect the presence of historical arrows;

[0020] If no historical arrow exists, the current arrow is stored as a historical arrow, and the following steps are repeated: photograph the road surface to obtain the arrow tip and arrow base;

[0021] If a historical arrow exists, the third position information of the current arrow and the fourth position information of the historical arrow are obtained, and the current arrow and the historical arrow are matched according to the third position information and the fourth position information.

[0022] Determine whether the arrow types of the current arrow and the historical arrow are the same after matching. If they are the same, associate the current arrow and the historical arrow.

[0023] Optionally, the step of matching the current arrow and the historical arrow based on the third location information and the fourth location information includes:

[0024] The distance between the current arrow and the historical arrow is determined based on the third location information and the fourth location information;

[0025] The current arrow and the historical arrow are matched based on the distance.

[0026] Optionally, after the step of associating the current arrow with the historical arrow, the method includes:

[0027] Obtain the cumulative number of updates to the fourth location information, and calculate the current location information based on the cumulative number of updates, the third location information, and the fourth location information;

[0028] The fourth location information is updated based on the current location information, and the cumulative number of updates is also updated.

[0029] Optionally, after the step of constructing the current arrow based on the arrow tip and the arrow base, the method further includes:

[0030] Compare the updated cumulative number of updates with the preset threshold number of updates;

[0031] If the cumulative number of updates after the update is greater than the preset number threshold, then the credibility of the current arrow is determined to meet the preset condition;

[0032] If the cumulative number of updates after the update is not greater than the preset number threshold, it is determined that the credibility of the current arrow does not meet the preset condition, and the historical arrow is updated to the current arrow, and the step of taking pictures of the road surface to obtain the arrow tip and arrow bottom of the road surface is executed again.

[0033] Furthermore, to achieve the above objectives, the present invention also provides a road directional arrow recognition device, the road directional arrow recognition device comprising:

[0034] The acquisition module is used to photograph the road surface and capture the tip and base of the arrows on the road surface;

[0035] The determination module is used to construct the current arrow based on the arrow tip and the arrow base. If the credibility of the current arrow meets the preset conditions, the current arrow is used as the road guide arrow.

[0036] Furthermore, the determining module further includes a construction module, which is used for:

[0037] Obtain the lane lines on the road surface, and determine the first position information of the arrow tip and the second position information of the arrow base;

[0038] Based on the lane lines, the first location information, and the second location information, the arrow tip and the arrow base are grouped;

[0039] Construct the current arrow based on the grouping results.

[0040] Furthermore, the building module is also used for:

[0041] Based on the lane lines, the first position information, and the second position information, determine whether the relative position of the arrow tip and the arrow base conforms to a preset rule;

[0042] Based on the first location information and the second location information, determine the line connecting the tip and the base of the arrow, and determine whether the line intersects with the lane line;

[0043] Based on the first location information and the second location information, determine whether the distance between the arrow tip and the arrow base is less than a preset distance threshold;

[0044] If it is determined that the relative position conforms to the preset rules, the connecting line does not intersect the lane line, and the distance is less than the preset distance threshold, then the arrow tip and the arrow base are grouped together.

[0045] Furthermore, the determining module further includes a detection module, the detection module being used for:

[0046] Detect the presence of historical arrows;

[0047] If no historical arrow exists, the current arrow is stored as a historical arrow, and the following steps are repeated: photograph the road surface to obtain the arrow tip and arrow base;

[0048] If a historical arrow exists, the third position information of the current arrow and the fourth position information of the historical arrow are obtained, and the current arrow and the historical arrow are matched according to the third position information and the fourth position information.

[0049] Determine whether the arrow types of the current arrow and the historical arrow are the same after matching. If they are the same, associate the current arrow and the historical arrow.

[0050] Furthermore, the determining module also includes a matching module, which is used for:

[0051] The distance between the current arrow and the historical arrow is determined based on the third location information and the fourth location information;

[0052] The current arrow and the historical arrow are matched based on the distance.

[0053] Furthermore, the determining module further includes an updating module, which is used to:

[0054] Obtain the cumulative number of updates to the fourth location information, and calculate the current location information based on the cumulative number of updates, the third location information, and the fourth location information;

[0055] The fourth location information is updated based on the current location information, and the cumulative number of updates is also updated.

[0056] Furthermore, the determining module also includes a judging module, which is used to:

[0057] Compare the updated cumulative number of updates with the preset threshold number of updates;

[0058] If the cumulative number of updates after the update is greater than the preset number threshold, then the credibility of the current arrow is determined to meet the preset condition;

[0059] If the cumulative number of updates after the update is not greater than the preset number threshold, it is determined that the credibility of the current arrow does not meet the preset condition, and the historical arrow is updated to the current arrow, and the step of taking pictures of the road surface to obtain the arrow tip and arrow bottom of the road surface is executed again.

[0060] In addition, to achieve the above objectives, the present invention also provides a road directional arrow recognition system, the road directional arrow recognition system comprising: a memory, a processor, and a road directional arrow recognition program stored in the memory and executable on the processor, wherein the road directional arrow recognition program, when executed by the processor, implements the steps of the road directional arrow recognition method as described above.

[0061] In addition, to achieve the above objectives, the present invention also provides a readable storage medium storing a road directional arrow recognition program, which, when executed by a processor, implements the steps of the road directional arrow recognition method as described above.

[0062] This invention proposes a method for recognizing road directional arrows. The method involves photographing the road surface to obtain the arrow tip and base. A current arrow is constructed based on the arrow tip and base. If the credibility of the current arrow meets a preset condition, it is designated as a road directional arrow. This invention, by constructing the current arrow from the arrow tip and base and identifying it as a road directional arrow when its credibility meets a preset condition, improves the accuracy of road directional arrow recognition even when the arrow is obscured and cannot be fully observed. Attached Figure Description

[0063] Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention;

[0064] Figure 2 This is a flowchart illustrating the first embodiment of the road directional arrow recognition method of the present invention;

[0065] Figure 3 This is a flowchart illustrating the second embodiment of the road surface guide arrow recognition method of the present invention;

[0066] Figure 4 This is a flowchart illustrating the third embodiment of the road surface guide arrow recognition method of the present invention;

[0067] Figure 5 This is a flowchart illustrating the fourth embodiment of the road surface guide arrow recognition method of the present invention;

[0068] Figure 6 This is a schematic diagram of the road surface guide arrow recognition device of the present invention.

[0069] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0070] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0071] like Figure 1 As shown, Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention.

[0072] The device in this embodiment of the invention can be a PC or a server.

[0073] like Figure 1 As shown, the device may include: a processor 1001, such as a CPU; a network interface 1004; a user interface 1003; a memory 1005; and a communication bus 1002. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0074] Those skilled in the art will understand that Figure 1 The device structure shown does not constitute a limitation on the device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0075] like Figure 1As shown, the memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a road directional arrow recognition program.

[0076] The operating system is a program that manages and controls the portable road directional arrow recognition system and its software resources, and supports the operation of the network communication module, user interface module, road directional arrow recognition program, and other programs or software; the network communication module is used to manage and control the network interface 1002; and the user interface module is used to manage and control the user interface 1003.

[0077] exist Figure 1 In the road directional arrow recognition system shown, the road directional arrow recognition system calls the road directional arrow recognition program stored in the memory 1005 through the processor 1001 and executes the operations in the various embodiments of the road directional arrow recognition method described below.

[0078] Based on the above hardware structure, an embodiment of the road directional arrow recognition method of the present invention is proposed.

[0079] Reference Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the road directional arrow recognition method of the present invention. The road directional arrow recognition method of this embodiment is applied to a road directional arrow recognition system, which can be applied to devices such as intelligent vehicles, intelligent terminals, and PC terminals. The method includes:

[0080] Step S10: Take a picture of the road surface to obtain the tip and base of the arrow on the road surface;

[0081] In this embodiment, during the driving process of the intelligent vehicle, the road directional arrow recognition system can control the camera in the intelligent vehicle to take road photos according to a preset time interval. The road directional arrow recognition system identifies the arrow tip and arrow base in the captured road photos. It is understood that the road directional arrow is usually composed of an arrow base and one or more arrow tips. In congested road sections, a part of the road directional arrow may be obscured by the vehicle, but when the arrow base or arrow tip is not obscured, the road directional arrow recognition system can identify the arrow tip and arrow base in the road photos.

[0082] Furthermore, when the tip or base of a road directional arrow is obscured by a vehicle, the road directional arrow recognition system can only obtain the unobscured tip or base. It needs to wait until the tip or base is unobscured before it can obtain the originally obscured tip or base. After obtaining the tip and base, the road directional arrow recognition system can perform the following steps.

[0083] Furthermore, after identifying the arrow tips and bases in the road surface photo, the road directional arrow recognition system will initially group the identified arrow tips and bases, grouping all arrow tips and bases that may belong to the same road surface arrow into the same group.

[0084] Furthermore, the road guidance arrow recognition system initially groups the identified arrow tips and bases. Since this is preliminary grouping, there will be many incorrect groupings. These incorrect groupings need to be separated, meaning the arrow tips within the incorrect groupings are stored as a separate group. Specifically, separating incorrect groupings involves the following three parts: A. The road guidance arrow recognition system identifies lane lines in the road photograph and obtains lane direction through lane line detection. The system then identifies the position information of the arrow tips and bases in each group and separates the incorrect groupings based on the lane direction and position information. For example, if the system identifies groups where the arrow tip representing a straight-ahead direction is located behind the arrow base along the lane direction, or where the arrow tip representing a right turn is located to the left of the arrow base, these are all incorrect groupings. Therefore, the arrow tips within these incorrect groupings are separated from the group and stored as a separate group. B. The road directional arrow recognition system determines the line connecting the arrow tip and the arrow base based on the position information of the arrow tip and the arrow base in each group, and judges whether the line connecting the arrow tip and the arrow base intersects with the lane line. If they intersect, it means that the arrow tip and the arrow base in the group belong to two different lanes. The arrow tip in the group is then separated from the group and stored as a separate group. C. The road directional arrow recognition system acquires the position / attitude information of the intelligent vehicle. Using IPM (Inverse Perspective Mapping), based on the position information of the arrow tip and base in each group and the position / attitude information of the intelligent vehicle, it determines the 3D position information of the arrow tip and base in each group. It then calculates the difference between the 3D position information of the arrow tip and base and compares this difference with a preset difference threshold. If the difference is greater than the preset threshold, the arrow tip in that group is separated from that group and stored as a separate group. It can be understood that the position information of the arrow tip and base is the 2D position information of the arrow tip and base in the road photograph, while the 3D position information of the arrow tip and base is the 3D position information of the arrow tip and base in the 3D map constructed by the intelligent vehicle.

[0085] Step S20: Construct a current arrow based on the arrow tip and the arrow base. If the credibility of the current arrow meets the preset conditions, then use the current arrow as a road guide arrow.

[0086] In this embodiment, the road directional arrow recognition system constructs the current arrow based on the arrow tip and the arrow base. After determining the current arrow, if the credibility of the current arrow meets the preset conditions, the road directional arrow recognition system uses the current arrow as the road directional arrow and releases the current arrow to the downstream for relevant processing.

[0087] Specifically, after acquiring the arrow tips and bases of the road surface, the road guidance arrow recognition system initially groups the arrow tips and bases. Correct groups exist where the arrow tips and bases do not fall under the conditions described in points A, B, and C. In this case, a correct group includes one arrow base and one or more arrow tips. Arrow tips from incorrect groups are separated into a separate group. The road guidance arrow recognition system analyzes each separately grouped arrow tip against the arrow bases in each correct group to determine if the conditions described in points A, B, and C exist. If not, the corresponding arrow tip is assigned to the correct group containing the corresponding arrow base, and together with the arrow bases and tips in the correct group, the current arrow is constructed. After determining the current arrow, the road guidance arrow recognition system checks if its confidence level meets preset conditions. If the confidence level meets the preset conditions, the current arrow is designated as a road guidance arrow, and its location information and arrow type are determined. The current arrow is then released to downstream processes.

[0088] The road directional arrow recognition system of this embodiment photographs the road surface to obtain the arrow tip and arrow base. Based on the arrow tip and arrow base, a current arrow is constructed. If the credibility of the current arrow meets a preset condition, it is designated as a road directional arrow. By photographing the road surface to obtain a road image, and then obtaining the arrow tip and arrow base from the image, the computational load required for photographing the road surface is small, and the frame rate of the road image is high. This results in a shorter time window for post-processing to suppress false detections, which is beneficial to improving the efficiency of road directional arrow recognition. By constructing the current arrow based on the arrow tip and arrow base, and determining the current arrow as a road directional arrow when its credibility meets the preset condition, the system can still recognize road directional arrows even when they are occluded and cannot be fully observed, thus improving the accuracy of road directional arrow recognition.

[0089] Further, refer to Figure 3Based on the first embodiment of the road directional arrow recognition method of the present invention, a second embodiment of the road directional arrow recognition method of the present invention is proposed.

[0090] The second embodiment of the road directional arrow recognition method differs from the first embodiment in that the step of constructing the current arrow based on the arrow tip and the arrow base includes:

[0091] Step S201: Obtain the lane lines on the road surface, and determine the first position information of the arrow tip and the second position information of the arrow base;

[0092] In this embodiment, the road guidance arrow recognition system identifies lane lines in a road surface photograph and determines a first position information of the arrow tip and a second position information of the arrow base based on the two-dimensional position information of the arrow tip and the two-dimensional position information of the arrow base in the road surface photograph, combined with the position / attitude information of the intelligent vehicle. It can be understood that the first position information includes the two-dimensional position information of the arrow tip in the road surface photograph and the three-dimensional position information of the arrow tip in the three-dimensional map constructed by the intelligent vehicle, and the second position information is the two-dimensional position information of the arrow base in the road surface photograph and the three-dimensional position information of the arrow base in the three-dimensional map constructed by the intelligent vehicle.

[0093] Step S202: Group the arrow tip and the arrow base according to the lane line, the first position information and the second position information;

[0094] In this embodiment, the road guide arrow recognition system determines whether the relative positions of the arrow tip and the arrow base conform to preset rules based on lane lines, first position information, and second position information; determines whether the line connecting the arrow tip and the arrow base intersects with the lane lines; and determines whether the distance between the arrow tip and the arrow base is less than a preset distance threshold, thereby determining to group the arrow tip and the arrow base.

[0095] Specifically, step S202 includes:

[0096] Step S2021: Based on the lane line, the first position information, and the second position information, determine whether the relative position of the arrow tip and the arrow base conforms to a preset rule;

[0097] In this step, the road guidance arrow recognition system obtains the lane direction through lane line detection. Based on the lane direction, the two-dimensional position information in the first position information of the arrow tip, and the two-dimensional position information in the second position information of the arrow base, the system determines the relative position of the arrow tip and the arrow base. After determining the arrow type corresponding to the arrow tip, the system further judges whether the relative position of the arrow tip and the arrow base conforms to preset rules based on the arrow type. For example, if the arrow tip of a straight-going arrow is located behind the arrow base along the lane direction, or the arrow tip of a right-turn arrow is located to the left of the arrow base, it can be determined that the relative position of the arrow tip and the arrow base does not conform to the preset rules. Conversely, if the arrow tip of a straight-going arrow is located in front of the arrow base along the lane direction, or the arrow tip of a right-turn arrow is located to the right of the arrow base, it can be determined that the relative position of the arrow tip and the arrow base conforms to the preset rules.

[0098] Step S2022: Determine the line connecting the tip of the arrow and the bottom of the arrow based on the first location information and the second location information, and determine whether the line intersects with the lane line;

[0099] In this step, the road directional arrow recognition system determines the relative position of the arrow tip on the road surface photo based on the first position information, determines the relative position of the arrow base on the road surface photo based on the second position information, and then determines the line connecting the arrow tip and the arrow base based on the relative positions of the arrow tip and the arrow base on the road surface photo, and determines whether the line intersects with the lane line.

[0100] Step S2023: Based on the first position information and the second position information, determine whether the distance between the arrow tip and the arrow base is less than a preset distance threshold;

[0101] In this step, the road guide arrow recognition system calculates the distance between the arrow tip and the arrow base based on the three-dimensional position information in the first position information and the three-dimensional position information in the second position information, and determines whether the distance is less than a preset distance threshold.

[0102] Step S2024: If it is determined that the relative position conforms to the preset rules, the connecting line does not intersect with the lane line, and the distance is less than the preset distance threshold, then the arrow tip and the arrow bottom are divided into the same group.

[0103] In this step, if the road guidance arrow recognition system determines that the relative positions of the arrow tip and the arrow base meet preset rules, the line connecting the arrow tip and the arrow base does not intersect the lane line, and the distance between the arrow tip and the arrow base is less than a preset distance threshold, then the arrow tip and the arrow base are grouped together. It should be noted that not all arrow tips can be matched with arrow bases and grouped together. If an arrow tip does not meet at least one of the above three conditions, it cannot be grouped with any arrow base. When no arrow base can be matched with a given arrow tip, that arrow tip is discarded.

[0104] Step S203: Construct the current arrow based on the grouping results.

[0105] In this step, after the road guidance arrow recognition system groups all arrow tips and arrow bases, it discards arrow tips that cannot be grouped and constructs the current arrow from the arrow tips and arrow bases that are in the same group. For example, suppose all arrow tips and arrow bases are divided into two groups. The first group includes an arrow tip of type right turn, an arrow tip of type straight, and an arrow base; the second group includes an arrow tip of type left turn and an arrow base. Correspondingly, the current arrow constructed from the arrow tips and arrow bases in the first group is a compound arrow of right turn and straight, and the current arrow constructed from the arrow tips and arrow bases in the second group is a single arrow of left turn.

[0106] The road directional arrow recognition system of this embodiment identifies lane lines in a road surface photograph and determines the first position information of the arrow tip and the second position information of the arrow base. Based on the lane lines, the first position information, and the second position information, the arrow tip and the arrow base are grouped. The current arrow is then constructed based on the grouping results. By grouping the arrow tip and the arrow base according to the lane lines, the first position information of the arrow tip, and the second position information of the arrow base to construct the current arrow, the system can construct the current arrow even if the road directional arrow portion in the road surface photograph is obscured, thus improving the accuracy of road directional arrow recognition when the road directional arrow portion is obscured.

[0107] Furthermore, such as Figure 4 As shown, based on the first and second embodiments of the road directional arrow recognition method of the present invention, a third embodiment of the road directional arrow recognition method of the present invention is proposed.

[0108] The third embodiment of the road directional arrow recognition method differs from the first and second embodiments in that, after the step of constructing the current arrow based on the arrow tip and the arrow base, it includes:

[0109] Step S204: Detect whether there are historical arrows;

[0110] In this embodiment, after constructing the current arrow, the road directional arrow recognition system detects whether there are historical arrows. It should be noted that the current arrow is the road directional arrow constructed from the arrow tip and arrow base in the current road surface photo acquired by the road directional arrow recognition system; the historical arrow is the road directional arrow constructed from the arrow tip and arrow base in historical road surface photos acquired by the road directional arrow recognition system before acquiring the current road surface photo. The road directional arrow recognition system can control the camera in the intelligent vehicle to take multiple road surface photos according to a preset time interval. The road surface photos need to be processed at the same time as they are acquired in order to identify the road directional arrow. Among the multiple road surface photos, the processed photos are historical road surface photos, and the photos being processed are current road surface photos.

[0111] Step S205: If no historical arrow exists, the current arrow is stored as a historical arrow, and the following steps are executed again: take a picture of the road surface to obtain the arrow tip and arrow base of the road surface;

[0112] In this embodiment, if the road directional arrow recognition system determines that no historical arrow exists, meaning that no other road surface photos were taken before the currently captured road surface photo, and therefore no historical arrow has been constructed, the road directional arrow recognition system stores the current arrow as a historical arrow and re-executes the following steps: taking a picture of the road surface to obtain the arrow tip and arrow base. This is to reconstruct a new current arrow, facilitating the subsequent determination of the road directional arrow by combining the current arrow and the historical arrow.

[0113] Step S206: If a historical arrow exists, obtain the third position information of the current arrow and the fourth position information of the historical arrow, and match the current arrow and the historical arrow according to the third position information and the fourth position information;

[0114] In this embodiment, after constructing the current arrow, if the road directional arrow recognition system determines that historical arrows exist, it acquires the third position information of each current arrow and the fourth position information of all historical arrows. It then matches the current arrow with the historical arrows based on the third and fourth position information, specifically matching the closest current arrow with the historical arrow. Essentially, after constructing the current arrow, the system acquires all historical arrows, identifies one historical arrow that matches the current arrow, and matches the current arrow with the matching historical arrow. It should be noted that the third position information includes the three-dimensional coordinate information corresponding to each arrow point in the current arrow, and the fourth position information includes the three-dimensional coordinate information corresponding to each arrow point in the historical arrows. The three-dimensional coordinate information represents the three-dimensional position of each arrow point in the three-dimensional map constructed by the intelligent vehicle.

[0115] Furthermore, when the current arrow cannot be matched with any historical arrow, the current arrow is stored in a new historical arrow for subsequent observations to match and update its position.

[0116] Furthermore, since each road surface photo may contain multiple arrow bases and multiple arrow tips, the road directional arrow recognition system may construct multiple current arrows. The above processing is performed on each current arrow, and a current arrow that meets the preset conditions is determined from all historical arrows. Each current arrow is then matched with its corresponding historical arrow that meets the conditions.

[0117] Specifically, step S206 includes:

[0118] Step S2061: Determine the distance between the current arrow and the historical arrow based on the third position information and the fourth position information;

[0119] Step S2062: Match the current arrow and the historical arrow according to the distance.

[0120] In steps S2061 to S2062, the road directional arrow recognition system determines the distance between the current arrow and historical arrows based on the third position information and the fourth position information, and matches the current arrow and historical arrows based on the distance. Specifically, for each current arrow, the road directional arrow recognition system determines the distance between the current arrow and each historical arrow based on the third position information of the current arrow and the fourth position information of all historical arrows, and then determines the historical arrow closest to the current arrow, and determines the arrow point closest to the current arrow in the historical arrows. After determining the arrow point, the road directional arrow recognition system calculates the distance between the arrow point and each current arrow, and then determines the current arrow closest to the arrow point, and determines whether the current arrow closest to the arrow point and the current arrow closest to the historical arrow where the arrow point is located are the same current arrow. If so, the current arrow and the historical arrow are matched.

[0121] For example, assuming there are three current arrows A, B, and C, and four historical arrows D, E, F, and G, for the current arrow A, the road guidance arrow recognition system determines the distances from the current arrow A to each of the four historical arrows D, E, F, and G based on the third position information of the current arrow A and the fourth position information of the four historical arrows D, E, F, and G. These four distances are then compared to determine the historical arrow closest to the current arrow A. Assuming that the distance from the current arrow A to the historical arrow D is determined to be the closest, the road guidance arrow recognition system identifies the arrow point d1 in historical arrow D that is closest to the current arrow A, and calculates... Calculate the distances from arrow point d1 to the three current arrows A, B, and C, and determine the current arrow closest to arrow point d1. Assuming that current arrow A is closest to arrow point d1, then determine that current arrow A is closest to historical arrow D, and that arrow point d1 in historical arrow D is also closest to current arrow A. That is, the current arrow closest to historical arrow D and the current arrow closest to arrow point d1 are the same current arrow A. Match current arrow A with historical arrow D. It is understandable that the same method is used to match historical arrows for current arrows B and C, which will not be elaborated here.

[0122] Step S207: Determine whether the arrow types of the matched current arrow and the historical arrow are the same. If they are the same, associate the current arrow and the historical arrow.

[0123] In this step, the road directional arrow recognition system matches the current arrow with historical arrows and determines whether the arrow types of the current arrow and historical arrows are the same. If they are the same, the current arrow and the historical arrow are associated.

[0124] For example, following the above example, after the road directional arrow recognition system matches the current arrow A and the historical arrow D, it determines whether the arrow types of the current arrow A and the historical arrow D are the same (e.g., whether the current arrow A and the historical arrow D are both straight single arrows, or whether the current arrow A and the historical arrow D are both straight and left-turn composite arrows, etc.). After determining that the arrow types of the current arrow A and the historical arrow D are the same, the current arrow and the historical arrow are associated, that is, it indicates that the current arrow A and the historical arrow D are the same road directional arrows that have been identified, such as the current arrow A and the historical arrow D being both straight and left-turn composite arrows.

[0125] Further, step S205 includes:

[0126] Step S206: Obtain the cumulative number of updates to the fourth location information, and calculate the current location information based on the cumulative number of updates, the third location information, and the fourth location information;

[0127] Step S207: Update the fourth position information according to the current arrow position information, and update the cumulative update count.

[0128] In steps S206 to S207, the road directional arrow recognition system associates the current arrow with historical arrows, obtains the cumulative update count of the fourth position information of the historical arrows associated with the current arrow, and calculates the position information of the current arrow based on the cumulative update count, the third position information, and the fourth position information; the road directional arrow recognition system updates the fourth position information based on the current position information and updates the cumulative update count corresponding to the historical arrows.

[0129] Specifically, the formula for calculating the current arrow's position information is: x' = (n*x0 + x1) / (n + 1), where x' is the current arrow's position information, n is the cumulative update count, x0 is the fourth position information of the historical arrow, and x1 is the third position information of the current arrow. The cumulative update count corresponding to the historical arrow is incremented by 1 for each of the above updates.

[0130] The road directional arrow recognition system of this embodiment acquires the third position information of the current arrow and the fourth position information of the historical arrows, and matches the current arrow and the historical arrows based on the third position information and the fourth position information; it then determines whether the arrow types of the matched current arrow and historical arrows are the same. If they are the same, the current arrow and the historical arrow are associated. By matching and associating the current arrow with the historical arrows, it is helpful to subsequently determine whether the credibility of the current arrow meets the preset conditions, thus improving the accuracy of road directional arrow recognition.

[0131] Furthermore, such as Figure 5As shown, based on the first to third embodiments of the road surface guide arrow recognition method of the present invention, a fourth embodiment of the road surface guide arrow recognition method of the present invention is proposed.

[0132] The fourth embodiment of the road directional arrow recognition method differs from the first to third embodiments in that step S30 includes:

[0133] Step S301: Compare the updated cumulative number of updates with the preset number of updates threshold;

[0134] In this embodiment, after updating the fourth position information based on the current arrow position information and updating the cumulative update count corresponding to the historical arrows, the road directional arrow recognition system compares the updated cumulative update count with a preset number threshold.

[0135] Step S302: If the cumulative number of updates after the update is greater than the preset number threshold, then it is determined that the credibility of the current arrow meets the preset condition.

[0136] In this embodiment, if the road directional arrow recognition system determines that the cumulative number of updates after the update is greater than a preset threshold, it determines that the credibility of the current arrow meets the preset condition and identifies the current arrow as the current road directional arrow. The road directional arrow recognition system then sends the arrow type and location information corresponding to the current arrow to the downstream for processing.

[0137] Step S303: If the cumulative number of updates after the update is not greater than the preset number threshold, then it is determined that the credibility of the current arrow does not meet the preset condition, the historical arrow is updated to the current arrow, and the step of acquiring the road surface photo and identifying the arrow tip and arrow bottom in the road surface photo is executed again.

[0138] In this embodiment, if the road directional arrow recognition system determines that the cumulative number of updates after the update is not greater than a preset threshold, it determines that the credibility of the current arrow does not meet the preset condition, updates the historical arrow to the current arrow, and re-executes the steps: taking a picture of the road surface to obtain the arrow tip and arrow base. It can be understood that when the credibility of the current arrow is determined not to meet the preset condition, the road directional arrow recognition system stores the current arrow as a historical arrow, re-acquires a new road surface photo, and executes the above steps until the credibility of the current arrow meets the preset condition. Then, it sends the arrow type and location information corresponding to the current arrow to the downstream for processing.

[0139] The road directional arrow recognition system in this embodiment compares the updated cumulative number of updates with a preset threshold. If the updated cumulative number of updates is greater than the preset threshold, the system determines that the credibility of the current arrow meets the preset condition and identifies the current arrow as the current road directional arrow. By using the cumulative update count of the historical arrows corresponding to the current arrow to determine whether the credibility of the current arrow meets the preset condition, the system avoids sending current arrows with unsatisfactory credibility to downstream applications, thus preventing recognition errors and improving the accuracy of road directional arrow recognition.

[0140] like Figure 6 As shown, the present invention also provides a road directional arrow recognition device. The road directional arrow recognition device of the present invention includes:

[0141] The acquisition module 101 is used to photograph the road surface and acquire the tip and base of the arrow on the road surface;

[0142] The determining module 102 is used to construct a current arrow based on the arrow tip and the arrow base. If the credibility of the current arrow meets a preset condition, the current arrow is used as a road guide arrow.

[0143] Furthermore, the determining module further includes a construction module, which is used for:

[0144] Obtain the lane lines on the road surface, and determine the first position information of the arrow tip and the second position information of the arrow base;

[0145] Based on the lane lines, the first location information, and the second location information, the arrow tip and the arrow base are grouped;

[0146] Construct the current arrow based on the grouping results.

[0147] Furthermore, the building module is also used for:

[0148] Based on the lane lines, the first position information, and the second position information, determine whether the relative position of the arrow tip and the arrow base conforms to a preset rule;

[0149] Based on the first location information and the second location information, determine the line connecting the tip and the base of the arrow, and determine whether the line intersects with the lane line;

[0150] Based on the first location information and the second location information, determine whether the distance between the arrow tip and the arrow base is less than a preset distance threshold;

[0151] If it is determined that the relative position conforms to the preset rules, the connecting line does not intersect the lane line, and the distance is less than the preset distance threshold, then the arrow tip and the arrow base are grouped together.

[0152] Furthermore, the determining module further includes a detection module, the detection module being used for:

[0153] Detect the presence of historical arrows;

[0154] If no historical arrow exists, the current arrow is stored as a historical arrow, and the following steps are repeated: photograph the road surface to obtain the arrow tip and arrow base;

[0155] If a historical arrow exists, the third position information of the current arrow and the fourth position information of the historical arrow are obtained, and the current arrow and the historical arrow are matched according to the third position information and the fourth position information.

[0156] Determine whether the arrow types of the current arrow and the historical arrow are the same after matching. If they are the same, associate the current arrow and the historical arrow.

[0157] Furthermore, the determining module also includes a matching module, which is used for:

[0158] The distance between the current arrow and the historical arrow is determined based on the third location information and the fourth location information;

[0159] The current arrow and the historical arrow are matched based on the distance.

[0160] Furthermore, the determining module further includes an updating module, which is used to:

[0161] Obtain the cumulative number of updates to the fourth location information, and calculate the current location information based on the cumulative number of updates, the third location information, and the fourth location information;

[0162] The fourth location information is updated based on the current location information, and the cumulative number of updates is also updated.

[0163] Furthermore, the determining module also includes a judging module, which is used to:

[0164] Compare the updated cumulative number of updates with the preset threshold number of updates;

[0165] If the cumulative number of updates after the update is greater than the preset number threshold, then the credibility of the current arrow is determined to meet the preset condition;

[0166] If the cumulative number of updates after the update is not greater than the preset number threshold, then it is determined that the credibility of the current arrow meets the preset condition, and the historical arrow is updated to the current arrow, and the step of taking pictures of the road surface to obtain the arrow tip and arrow bottom of the road surface is executed again.

[0167] The present invention also provides a road directional arrow recognition system.

[0168] The road directional arrow recognition system of the present invention includes: a memory, a processor, and a road directional arrow recognition program stored in the memory and executable on the processor. When the road directional arrow recognition program is executed by the processor, it implements the steps of the road directional arrow recognition method as described above.

[0169] The method implemented when the road directional arrow recognition program running on the processor is executed can be referred to in various embodiments of the road directional arrow recognition method of the present invention, and will not be repeated here.

[0170] The present invention also provides a readable storage medium.

[0171] The present invention provides a road directional arrow recognition program stored on a readable storage medium, which, when executed by a processor, implements the steps of the road directional arrow recognition method as described above.

[0172] The method implemented when the road directional arrow recognition program running on the processor is executed can be referred to in various embodiments of the road directional arrow recognition method of the present invention, and will not be repeated here.

[0173] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0174] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0175] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0176] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for recognizing road directional arrows, characterized in that, The road directional arrow recognition method includes the following steps: Photograph the road surface to capture the tip and base of the arrows on the road surface; The current arrow is constructed based on the arrow tip and the arrow base. If the credibility of the current arrow meets the preset conditions, the current arrow is used as the road guide arrow. Following the step of constructing the current arrow based on the arrow tip and the arrow base, the method includes: Detect the presence of historical arrows; If a historical arrow exists, the third position information of the current arrow and the fourth position information of the historical arrow are obtained, and the current arrow and the historical arrow are matched according to the third position information and the fourth position information. Determine whether the arrow types of the current arrow and the historical arrow are the same after matching. If they are the same, associate the current arrow and the historical arrow. Obtain the cumulative number of updates to the fourth location information, and calculate the current location information based on the cumulative number of updates, the third location information, and the fourth location information; The fourth location information is updated based on the current location information, and the cumulative update count is updated. The cumulative update count is used to help determine whether the credibility of the current arrow meets the preset conditions.

2. The road directional arrow recognition method as described in claim 1, characterized in that, The step of constructing the current arrow based on the arrow tip and the arrow base includes: Obtain the lane lines on the road surface, and determine the first position information of the arrow tip and the second position information of the arrow base; Based on the lane lines, the first location information, and the second location information, the arrow tip and the arrow base are grouped; Construct the current arrow based on the grouping results.

3. The road directional arrow recognition method as described in claim 2, characterized in that, The step of grouping the arrow tip and the arrow base according to the lane line, the first position information, and the second position information includes: Based on the lane lines, the first position information, and the second position information, determine whether the relative position of the arrow tip and the arrow base conforms to a preset rule; Based on the first location information and the second location information, determine the line connecting the tip and the base of the arrow, and determine whether the line intersects with the lane line; Based on the first location information and the second location information, determine whether the distance between the arrow tip and the arrow base is less than a preset distance threshold; If it is determined that the relative position conforms to the preset rules, the connecting line does not intersect the lane line, and the distance is less than the preset distance threshold, then the arrow tip and the arrow base are grouped together.

4. The road directional arrow recognition method as described in claim 1, characterized in that, Following the step of detecting the existence of historical arrows, the method further includes: If no historical arrow exists, the current arrow is stored as a historical arrow, and the following steps are repeated: photograph the road surface to obtain the arrow tip and arrow base.

5. The road directional arrow recognition method as described in claim 1, characterized in that, The step of matching the current arrow and the historical arrow based on the third location information and the fourth location information includes: The distance between the current arrow and the historical arrow is determined based on the third location information and the fourth location information; The current arrow and the historical arrow are matched based on the distance.

6. The road directional arrow recognition method as described in claim 1, characterized in that, Following the step of constructing the current arrow based on the arrow tip and the arrow base, the method further includes: Compare the updated cumulative number of updates with the preset threshold number of updates; If the cumulative number of updates after the update is greater than the preset number threshold, then the credibility of the current arrow is determined to meet the preset condition; If the cumulative number of updates after the update is not greater than the preset number threshold, it is determined that the credibility of the current arrow does not meet the preset condition, and the historical arrow is updated to the current arrow, and the step of taking pictures of the road surface to obtain the arrow tip and arrow bottom of the road surface is executed again.

7. A road directional arrow recognition device, characterized in that, The road surface guide arrow recognition device includes: The acquisition module is used to photograph the road surface and capture the tip and base of the arrows on the road surface; The determination module is used to construct the current arrow based on the arrow tip and the arrow base. If the confidence level of the current arrow meets the preset conditions, the current arrow is used as the road guide arrow. The determining module is further configured to detect whether a historical arrow exists after the step of constructing the current arrow based on the arrow tip and the arrow base; If a historical arrow exists, the third position information of the current arrow and the fourth position information of the historical arrow are obtained, and the current arrow and the historical arrow are matched according to the third position information and the fourth position information. Determine whether the arrow types of the current arrow and the historical arrow are the same after matching. If they are the same, associate the current arrow and the historical arrow. Obtain the cumulative number of updates to the fourth location information, and calculate the current location information based on the cumulative number of updates, the third location information, and the fourth location information; The fourth location information is updated based on the current location information, and the cumulative update count is updated. The cumulative update count is used to help determine whether the credibility of the current arrow meets the preset conditions.

8. A road directional arrow recognition system, characterized in that, The road directional arrow recognition system includes: a memory, a processor, and a road directional arrow recognition program stored in the memory and executable on the processor. When the road directional arrow recognition program is executed by the processor, it implements the steps of the road directional arrow recognition method as described in any one of claims 1 to 6.

9. A readable storage medium, characterized in that, The readable storage medium stores a road directional arrow recognition program, which, when executed by a processor, implements the steps of the road directional arrow recognition method as described in any one of claims 1 to 6.

Images