Detection method and detection device for special traffic lights

Abstract

The disclosed embodiments provide a method, apparatus, storage medium, and electronic device for detecting special traffic lights. The detection method includes determining the projection position of the special traffic light and the number of sub-lights within the special traffic light; obtaining information about the special traffic light's bounding box; and obtaining color information for each sub-light within the bounding box. The disclosed embodiments can separate the conventional method of directly detecting color in ordinary traffic light perception into a method of combining detection box type with color classification, providing greater redundancy and offering a universally applicable solution for different types of special traffic lights. This effectively addresses the current problem of being unable to detect the status of special traffic lights and offers a high degree of detection accuracy for special traffic lights.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the technical field of traffic light recognition, and more particularly to a detection method, device, storage medium, and electronic device for special traffic lights. Background Art

[0002] Traffic light recognition is an essential component of autonomous driving technology. Currently, research on traffic light recognition methods in the autonomous driving field falls into two main categories: camera-based methods (i.e., visual perception detection methods) and vehicle-to-everything (V2X)-based methods (i.e., methods that rely on vehicle-to-traffic light communication). Camera-based methods primarily rely on training deep learning networks on labeled traffic light data to learn image features for recognition. For standard traffic lights, which typically have only four categories—red, green, yellow, and black (flashing)—this method is effective for identifying standard traffic lights while ensuring sufficient data volume. However, for special traffic lights, the number of sub-lights and the lighting pattern are uncertain, resulting in variable and uncertain classification, and the color output is undefined. Therefore, directly detecting the color state of special traffic lights using deep learning networks is not feasible. V2X-based methods require communication equipment installed on both the traffic light device and the vehicle. When passing through an intersection, the vehicle receives real-time communication signals from the traffic light device and analyzes them to obtain the latest traffic light status. This method is safer and more responsive than camera recognition, but it is more dependent on the popularity of road test communication equipment. Currently, its popularity is still relatively low and is only used in some autonomous driving demonstration operation areas.

[0003] Currently, most traffic light identification methods on the road mainly detect ordinary traffic lights, where only one light in a frame is illuminated at a time. There are no effective detection methods or devices for some special traffic lights, making effective identification impossible. Special traffic lights are lights with multiple sub-lights of the same or different colors within a frame, which can illuminate simultaneously. Summary of the Invention

[0004] In order to solve the problem that the visual perception module in the existing autonomous driving can only detect ordinary traffic lights but cannot detect special traffic lights, the purpose of the embodiments of the present invention is to provide a detection method, device, storage medium and electronic device for special traffic light status, so as to solve the above problems in the prior art.

[0005] In order to solve the above technical problems, the embodiments of the present disclosure adopt the following technical solutions:

[0006] A detection method for a special traffic light comprises: determining the projection position of the special traffic light and the number of sub-lights in the special traffic light; obtaining information of an external frame of the special traffic light; and obtaining color information of each sub-light in the external frame.

[0007] In some embodiments, determining the projection position of the special traffic light includes: obtaining a first position of the special traffic light in the world coordinate system; based on the first position, obtaining a second position of the special traffic light in the vehicle body coordinate system; based on the second position, obtaining the projection position of the special traffic light in the image coordinate system.

[0008] In some embodiments, obtaining the information of the external frame of the special traffic light includes: obtaining feature information of the detection frame of the special traffic light through a deep learning network; obtaining feature information of the projection position frame of the special traffic light; matching the detection frame and the projection position frame to obtain information of the external frame.

[0009] In some embodiments, the feature information includes at least type information and location information.

[0010] In some embodiments, matching the detection frame and the projection position frame to obtain information of the circumscribed frame includes: obtaining the first center point coordinates of the detection frame; obtaining the second center point coordinates of the projection position frame; calculating the Euclidean distance based on the first center point coordinates and the second center point coordinates; and obtaining information of the circumscribed frame based on the Euclidean distance.

[0011] In some embodiments, obtaining the color information of each sub-lamp in the external frame includes: dividing the external frame based on the number of sub-lamp; obtaining the color information of each sub-lamp in the divided external frame.

[0012] In some embodiments, obtaining the color information of each sub-lamp in the segmented external frame includes: converting the segmented image of each sub-lamp from the RGB space to the HSV space; and identifying and obtaining the color of the sub-lamp in the HSV space.

[0013] The present disclosure also provides a detection device for special traffic lights, which includes: a determination module for determining the projection position of the special traffic light and the number of sub-lights in the special traffic light; an external frame acquisition module for acquiring information of the external frame of the special traffic light; and a color acquisition module for acquiring color information of each sub-light in the external frame.

[0014] The present disclosure also provides a storage medium storing a computer program, wherein the computer program implements the steps of any of the above methods when executed by a processor.

[0015] The present disclosure also provides an electronic device, comprising at least a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any of the above methods when executing the computer program on the memory.

[0016] The beneficial effects of the embodiments of the present disclosure are that: the embodiments of the present disclosure can split the method of directly detecting color in the usual ordinary traffic light perception into a method of detection frame type plus color classification, providing better redundancy, and providing a universally applicable solution for different types of special traffic lights, effectively solving the current problem of not being able to detect the status of special traffic lights, and having a high detection accuracy for special traffic lights. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0018] Figure 1 Schematic diagram of the steps of a method for detecting a special traffic light according to an embodiment of the present disclosure;

[0019] Figure 2 Schematic diagram of the steps of a method for detecting a special traffic light according to an embodiment of the present disclosure;

[0020] Figure 3 Schematic diagram of the steps of a method for detecting a special traffic light according to an embodiment of the present disclosure;

[0021] Figure 4 Schematic diagram of the steps of a method for detecting special traffic lights according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0022] Various aspects and features of the present disclosure are described herein with reference to the accompanying drawings.

[0023] It should be understood that various modifications may be made to the embodiments disclosed herein. Therefore, the above description should not be considered as limiting, but merely as an example of an embodiment. Other modifications within the scope and spirit of the present disclosure will occur to those skilled in the art.

[0024] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the general description of the present disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the present disclosure.

[0025] These and other characteristics of the present disclosure will become apparent from the following description of a preferred form of embodiment given as a non-limiting example with reference to the accompanying drawings.

[0026] It should also be understood that although the present disclosure has been described with reference to certain specific examples, those skilled in the art will be able to realize many other equivalent forms of the present disclosure that have the characteristics recited in the claims and are therefore within the scope of protection defined thereby.

[0027] The above and other aspects, features and advantages of the present disclosure will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings.

[0028] Specific embodiments of the present disclosure will be described hereinafter with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of the present disclosure, which may be implemented in a variety of ways. Well-known and / or repetitive functions and structures are not described in detail to avoid obscuring the present disclosure with unnecessary or redundant detail. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but rather serve merely as a basis and representative basis for teaching those skilled in the art to variously employ the present disclosure with substantially any suitable detailed structure.

[0029] This description may use the phrases "in one embodiment," "in another embodiment," "in a further embodiment," or "in other embodiments," each of which may refer to one or more of the same or different embodiments according to the present disclosure.

[0030] The first embodiment of the present disclosure is used in the field of visual perception detection in autonomous driving scenarios, and specifically relates to a detection method for special traffic light states that realizes multi-module fusion in autonomous driving. The special traffic light here refers to a light frame with multiple sub-lights of the same or different colors, and these sub-lights can light up at the same time or at different times.

[0031] The disclosed embodiment can utilize the mutual integration of the high-precision map module, the traffic light detection frame module and the traffic light color classification module in the vehicle system to realize the detection of special traffic lights.

[0032] like Figure 1 As shown, the detection method includes the following steps:

[0033] S101, determining the projection position of the special traffic light and the number of sub-lights in the special traffic light.

[0034] In this step, the projection position of the special traffic light and the number of sub-lights in the special traffic light are determined. The position information of the special traffic light here can be pre-recorded in the database of the high-definition map module. Based on the high-definition map module, the position information of the special traffic light in the high-definition map is first obtained and further converted into the projection position. At the same time, the number of sub-lights in the special traffic light is obtained based on the high-definition map module. Specifically, Figure 2 As shown, the following steps are included:

[0035] S201: Obtain a first position of the special traffic light in the world coordinate system.

[0036] In this step, the first position of the special traffic light in the world coordinate system is obtained. Specifically, based on, for example, the high-precision map module, the coordinates of the first position of the special traffic light can be obtained in the world coordinate system of the high-definition map. Here, for example, the coordinates of the first position of the special traffic light in the world coordinate system can be obtained by the high-precision map module as (x g ,y g ,z g ).

[0037] S202: Based on the first position, obtain a second position of the special traffic light in the vehicle body coordinate system.

[0038] After the first position of the special traffic light in the world coordinate system is obtained in step S201, in this step, the second position of the special traffic light in the vehicle body coordinate system is obtained based on the first position. Specifically, when the coordinates of the first position of the special traffic light in the world coordinate system are (x g ,y g ,z g ), the coordinates of the first position can be transformed by the external parameter matrix LW to obtain the coordinates of the second position of the special traffic light in the vehicle body coordinate system (x c ,y c ,z c ), the conversion formula here is:

[0039]

[0040] The extrinsic parameter matrix LW here can be obtained by extrinsic parameter calibration of the vehicle, and R and t here represent the rotation matrix and the translation matrix respectively.

[0041] S203: Based on the second position, obtain the projection position of the special traffic light in the image coordinate system.

[0042] After obtaining the second position of the special traffic light in the vehicle body coordinate system based on the first position through the above step S202, in this step, the projection position of the special traffic light in the image coordinate system is obtained based on the second position.

[0043] Specifically, considering that the coordinates of the second position of the special traffic light in the vehicle body coordinate system are (x c ,y c ,z c ), the coordinates of the second position can be converted into the projection position (x, y) of the special traffic light in the image coordinate system by, for example, the intrinsic parameter matrix K of the camera device on the vehicle. The conversion formula here is:

[0044]

[0045] Here, K is the intrinsic parameter matrix of the camera device on the vehicle, f is the focal length of the camera device, and (Cx, Cy) is the coordinate of the optical center of the camera device in the image plane, usually corresponding to the coordinate (0, 0).

[0046] In this way, through the above two conversion formulas, the coordinates of the first position of the special traffic light in the world coordinate system can be converted to the coordinates of its projection position in the image coordinate system, so that the accurate position of the special traffic light can be obtained.

[0047] In addition, since a special traffic light may include one or more sub-lights, in step S101, the high-definition map provided by the high-precision map module can also identify and determine the number n of sub-lights in the special traffic light. The identification method here is not limited here.

[0048] Specifically, in one embodiment, for example, the following information of each special traffic light can be obtained through the high-precision map module, namely {x1, y1, x2, y2, n}, where (x1, y1) represents the upper left corner coordinate of the projection position of the special traffic light, (x2, y2) represents the lower right corner coordinate of the projection position of the special traffic light, and n is the number of sub-lights in the special traffic light.

[0049] S102: Obtain information about the external frame of the special traffic light.

[0050] After determining the projection position of the special traffic light and the number of sub-lights in the special traffic light in step S101, in this step, the information of the external frame of the special traffic light is obtained. The external frame here is the frame area in the special traffic light used for information identification. Specifically, Figure 3 As shown, the following steps are included:

[0051] S301, obtaining feature information of the detection frame of the special traffic light through a deep learning network.

[0052] In this step, a deep learning network is used to obtain feature information for the detection frame of the special traffic light. The detection frame is the area of ​​the special traffic light that includes the sub-lights. This feature information can include, for example, the detection frame's location and type. In this step, detection is performed solely based on information such as the detection frame's type and location, without distinguishing specific colors.

[0053] Specifically, considering the actual shape of the special traffic light and to accommodate various special traffic light shapes, a detection frame module based on the special traffic light uses a deep learning network to obtain the type of the detection frame. The type can be multi-dimensional, such as differentiating the detection frame into a vertical frame or a horizontal frame, and other shapes are also possible. The detection frame module employed herein includes at least hardware such as a camera device and software such as a deep learning network.

[0054] Furthermore, new shapes for special traffic lights can be directly added to the deep learning network. This allows the traffic light detection frame module to output only the type of the special traffic light's bounding box instead of the color, providing good redundancy and making detection easier.

[0055] Furthermore, the detection frame module based on the special traffic light detects and obtains the position information of the detection frame of the special traffic light through a deep learning network. Specifically, a deep learning detection network such as a Yolov4 network can be used to detect the position information of the detection frame of the special traffic light, thereby obtaining feature information of the detection frame of the special traffic light.

[0056] Specifically, in one embodiment, accurate feature information of the detection frame of the special traffic light can be obtained, such as {xi1, yi1, xi2, yi2, Ci}, where (xi1, yi1) is the upper left coordinate of the i-th detection frame, (xi2, yi2) is the lower right coordinate of the i-th detection frame, and Ci is the type of the i-th special traffic light.

[0057] S302: Acquire feature information of the projection position frame of the special traffic light.

[0058] After obtaining the feature information of the detection frame of the special traffic light through the deep learning network in the above step S301, in this step, the feature information of the projection position frame of the special traffic light is obtained through, for example, a high-precision map module on the vehicle.

[0059] Considering that there may be misdetection in the detection result of the detection frame of the special traffic light in step S301, the high-precision map module and the detection module are fused and matched in this step to ensure the correctness of the result.

[0060] S303: Match the detection frame with the projection position frame to obtain information of the circumscribed frame.

[0061] After obtaining the feature information of the detection frame of the special traffic light through the deep learning network in the above step S301 and obtaining the feature information of the projection position frame of the special traffic light through the high-precision map module in the above step S302, in this step, the detection frame and the projection position frame are matched to obtain the information of the external frame.

[0062] Specifically, the frame information obtained by the high-precision map module and the traffic light detection frame module is fused and matched, filtering out possible false detections and retaining the final matched external frame information. Any matching method can be used here, such as the Euclidean distance method.

[0063] Specifically, for example, after obtaining the position of the detection frame of the special traffic light according to step S301, the center point coordinate x of the detection frame can be obtained. cen ,y cen :

[0064]

[0065] Further, according to the above step S302, the center point coordinates x of each projection position frame are calculated. ijc ,y ijc :

[0066]

[0067] In this way, the Euclidean distance Li between the center point of the detection frame and the center point of each projection position frame can be calculated:

[0068]

[0069] In this step, the Euclidean distance Li calculated above is used as the feature value of Hungarian matching. By setting a threshold of the Euclidean distance, possible false detections can be filtered out, and the correct detection results of the bounding box can be retained, thereby obtaining the information of the bounding box.

[0070] In this way, the projection position of the special traffic light obtained by the high-precision map module and the position of the detection frame of the special traffic light are matched using the Euclidean distance feature, filtering out false detection frames while making the detection more accurate. At the same time, in this step, due to the combination of the projection position of the special traffic light obtained by the high-precision map module, the false recognition rate in the detection is greatly reduced.

[0071] S103: Acquire color information of each sub-lamp in the external frame.

[0072] After obtaining the information of the outer frame of the special traffic light in step S102, in this step, the color information of each sub-light in the outer frame is obtained. Here, the color information of at least one sub-light in the outer frame of the special traffic light is identified. Specifically, Figure 4 As shown, the following steps are included:

[0073] S401: Divide the external frame based on the number of sub-lamps.

[0074] In this step, the external frame is first segmented based on the number of sub-lights. Specifically, based on the number of sub-lights provided by the high-precision map module, different segmentation methods can be used to segment the external frame of different types (such as vertical frames or horizontal frames) based on the number and distribution of the sub-lights.

[0075] For example, based on the number of sub-lights n given by the high-precision map module and the position and type Ci of the bounding box of the matched special traffic light obtained in step S102, the specific position of each sub-light in the bounding box can be calculated. Here, it is assumed that the type of the special traffic light detected is a vertical light (it may also be horizontal in actual conditions), and the height h of the sub-light is:

[0076]

[0077] The upper left corner coordinates and lower right corner coordinates of each sub-lamp are:

[0078] x iz1 ,y iz1 =x i1 ,y i1 +m*h

[0079] x iz2 ,y iz2 =x i2 ,y i2 +(m+1)*h

[0080] Where m is a positive integer between 0 and (n-1).

[0081] In this way, n images of small lamps can be segmented from the image of the circumscribed frame according to the coordinate values, thereby achieving segmentation of the circumscribed frame.

[0082] S402: Acquire color information of each sub-lamp in the divided external frame.

[0083] After the external frame is segmented based on the number of sub-lights in step S401, this step obtains the color information of each sub-light within the segmented external frame. Specifically, the image of each segmented sub-light can be converted from RGB space to HSV space based on the traffic light color classification module, and the color can be identified using traditional methods. The image of each sub-light is converted from RGB space to HSV space (for example, including features such as hue / saturation / lightness color space) mainly because color identification is easier in HSV space. Finally, the color results of each sub-light in the special traffic light are output to the back-end control system.

[0084] The disclosed embodiment integrates the high-precision map module, the traffic light detection frame module and the traffic light color classification module, thereby splitting the method of directly detecting color in ordinary traffic light perception into a method of detection frame type plus color classification, providing better redundancy and a universally applicable solution for different types of special traffic lights, effectively solving the current problem of being unable to detect the status of special traffic lights, and having a high detection accuracy for special traffic lights.

[0085] The second embodiment of the present disclosure relates to a detection device for special traffic light status that realizes multi-module fusion in autonomous driving. The special traffic light here refers to a light frame with multiple sub-lights of the same or different colors, and these sub-lights can light up at the same time or at different times.

[0086] The detection device includes a determination module, an external frame acquisition module, and a color acquisition module. Specifically:

[0087] The determination module is used to determine the projection position of the special traffic light and the number of sub-lights in the special traffic light.

[0088] Specifically, the determination module includes the following parts:

[0089] The first position acquisition unit is used to acquire the first position of the special traffic light in the world coordinate system.

[0090] The second position acquisition unit is used to acquire a second position of the special traffic light in the vehicle body coordinate system based on the first position.

[0091] A projection position acquisition unit is used to acquire the projection position of the special traffic light in the image coordinate system based on the second position.

[0092] The external frame acquisition module is used to obtain information about the external frame of the special traffic light.

[0093] The external frame acquisition module includes the following parts:

[0094] The first feature information acquisition unit is used to obtain feature information of the detection frame of the special traffic light through a deep learning network.

[0095] The second feature information acquisition unit is used to obtain the feature information of the projection position frame of the special traffic light through the high-precision map module.

[0096] The matching unit is used to match the detection frame with the projection position frame to obtain information of the circumscribed frame.

[0097] The color acquisition module is used to acquire the color information of each sub-lamp in the external frame.

[0098] The color acquisition module includes the following steps:

[0099] The segmentation unit is configured to segment the external frame based on the number of sub-lamps.

[0100] The color acquisition unit is used to acquire the color information of each sub-lamp in the divided external frame.

[0101] The disclosed embodiment splits the method of directly detecting color in the usual ordinary traffic light perception into a method of detection frame type plus color classification, which provides better redundancy and provides a universally applicable solution for different types of special traffic lights. It effectively solves the current problem of being unable to detect the status of special traffic lights and has a high detection accuracy for special traffic lights.

[0102] A third embodiment of the present disclosure provides a storage medium, which is a computer-readable medium and stores a computer program. When the computer program is executed by a processor, the method provided by the first and third embodiments of the present disclosure is implemented, including the following steps S11 to S13:

[0103] S11, determining the projection position of the special traffic light and the number of sub-lights in the special traffic light;

[0104] S12, obtaining information about the external frame of the special traffic light;

[0105] S13, obtaining color information of each sub-lamp in the external frame.

[0106] Furthermore, when the computer program is executed by a processor, the first embodiment of the present disclosure and other methods provided by the first embodiment are implemented.

[0107] The disclosed embodiment splits the method of directly detecting color in the usual ordinary traffic light perception into a method of detection frame type plus color classification, which provides better redundancy and provides a universally applicable solution for different types of special traffic lights. It effectively solves the current problem of being unable to detect the status of special traffic lights and has a high detection accuracy for special traffic lights.

[0108] A fourth embodiment of the present disclosure provides an electronic device, which includes at least a memory and a processor. The memory stores a computer program, and the processor implements the method provided by any embodiment of the present disclosure when executing the computer program in the memory. For example, the electronic device computer program steps S21 to S23 are as follows:

[0109] S21, determining the projection position of the special traffic light and the number of sub-lights in the special traffic light;

[0110] S22, obtaining information about the external frame of the special traffic light;

[0111] S23, obtaining color information of each sub-lamp in the external frame.

[0112] Furthermore, the processor also executes the computer programs in the first and second embodiments above.

[0113] The disclosed embodiment splits the method of directly detecting color in the usual ordinary traffic light perception into a method of detection frame type plus color classification, which provides better redundancy and provides a universally applicable solution for different types of special traffic lights. It effectively solves the current problem of being unable to detect the status of special traffic lights and has a high detection accuracy for special traffic lights.

[0114] The storage medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0115] The storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains at least two Internet Protocol addresses; sends a node evaluation request including at least two Internet Protocol addresses to a node evaluation device, wherein the node evaluation device selects an Internet Protocol address from the at least two Internet Protocol addresses and returns it; receives the Internet Protocol address returned by the node evaluation device; wherein the obtained Internet Protocol address indicates an edge node in a content distribution network.

[0116] Alternatively, the storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: receive a node evaluation request including at least two Internet Protocol addresses; select an Internet Protocol address from the at least two Internet Protocol addresses; and return the selected Internet Protocol address; wherein the received Internet Protocol address indicates an edge node in a content distribution network.

[0117] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including, but not limited to, object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the passenger computer, partially on the passenger computer, as a stand-alone software package, partially on the passenger computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the passenger computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0118] It should be noted that the storage medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or component. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any storage medium other than a computer-readable storage medium that can transmit, propagate, or convey a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the storage medium may be conveyed using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), or any suitable combination thereof.

[0119] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0120] The units involved in the embodiments described in this disclosure may be implemented in software or hardware, wherein the name of a unit does not necessarily limit the unit itself.

[0121] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0122] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0123] The above description is merely a preferred embodiment of the present disclosure and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also includes other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this disclosure.

[0124] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details have been included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0125] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

[0126] The above describes in detail multiple embodiments of the present disclosure, but the present disclosure is not limited to these specific embodiments. Those skilled in the art can make various variations and modifications to the embodiments based on the concepts of the present disclosure, and these variations and modifications should all fall within the scope of protection claimed by the present disclosure.

Claims

1. A method for detecting a special traffic light, wherein the special traffic light is a light frame having multiple sub-lights of different colors, the sub-lights being lit simultaneously or at different times, characterized in that: include: Determining the projection position of the special traffic light and the number of sub-lights in the special traffic light, wherein determining the projection position of the special traffic light includes: obtaining position information of the special traffic light in a high-definition map, and obtaining the projection position of the special traffic light based on the position information; Obtaining information about the external frame of the special traffic light, including: obtaining feature information of a detection frame of the special traffic light through a deep learning network, obtaining feature information of a projection position frame of the special traffic light, matching and fusing the detection frame and the projection position frame to obtain information about the external frame; Acquire color information of each sub-lamp in the external frame; The acquiring the color information of each sub-lamp in the external frame includes: dividing the external frame based on the number of sub-lamp; Identify the color information of each sub-lamp in the divided external frame; The color information of each sub-lamp in the external frame after identification includes: Convert the segmented image of each sub-light from the RGB space to the HSV space; The color of the sub-lamp is identified and obtained in the HSV space.

2. The detection method according to claim 1, wherein Determining the projection position of the special traffic light includes: Obtaining the first position of the special traffic light in the world coordinate system; Based on the first position, obtaining a second position of the special traffic light in a vehicle body coordinate system; Based on the second position, the projection position of the special traffic light in the image coordinate system is obtained.

3. The detection method according to claim 1, wherein The characteristic information includes at least type information and location information.

4. The detection method according to claim 1, wherein The matching of the detection frame and the projection position frame to obtain information of the circumscribed frame includes: Obtaining the coordinates of a first center point of the detection frame; Obtaining the coordinates of the second center point of the projection position frame; Calculating the Euclidean distance based on the first center point coordinates and the second center point coordinates; Based on the Euclidean distance, information about the bounding box is obtained.

5. A detection device for a special traffic light, wherein the special traffic light is a light frame having multiple sub-lights of different colors, the sub-lights lighting up simultaneously or at different times, characterized in that: include: a determination module, configured to determine a projection position of the special traffic light and the number of sub-lights in the special traffic light, wherein determining the projection position of the special traffic light comprises: obtaining position information of the special traffic light in a high-definition map, and obtaining the projection position of the special traffic light according to the position information; an external frame acquisition module, configured to acquire information about the external frame of the special traffic light, including: acquiring feature information of a detection frame of the special traffic light through a deep learning network, acquiring feature information of a projection position frame of the special traffic light, matching and fusing the detection frame and the projection position frame, and acquiring information about the external frame; A color acquisition module, configured to acquire color information of each sub-lamp in the external frame; The acquiring the color information of each sub-lamp in the external frame includes: dividing the external frame based on the number of sub-lamp; Identify the color information of each sub-lamp in the divided external frame; The color information of each sub-lamp in the external frame after identification includes: Convert the segmented image of each sub-light from the RGB space to the HSV space; The color of the sub-lamp is identified and obtained in the HSV space.

6. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

7. An electronic device comprising at least a memory and a processor, wherein the memory stores a computer program, wherein: The processor implements the steps of the method of any one of claims 1 to 4 when executing the computer program on the memory.

Images