CN118212611A - Signal identification method and device of indicator lamp, vehicle-mounted equipment and storage medium - Google Patents

Abstract

The embodiment of the application is suitable for the technical field of traffic, and provides a signal identification method and device of an indicator lamp, vehicle-mounted equipment and a storage medium, wherein the method comprises the following steps: acquiring left-eye images and right-eye images of target indicator lamps at the intersection; determining a first distance between the vehicle and the target indicator lamp according to the left eye image and the right eye image; if the first distance is greater than the preset distance, repeating the steps of acquiring left-eye images and right-eye images of the target indicator lights at the intersection and obtaining the first distance; and if the first distance is smaller than or equal to the preset distance, identifying the target indicator lamp according to the left-eye image or the right-eye image to obtain a traffic signal of the target indicator lamp. By adopting the method, the accuracy of signal identification of traffic lights at the intersections can be improved, and the consumed computing resources can be reduced.

Description

Signal identification method and device of indicator lamp, vehicle-mounted equipment and storage medium

Technical Field

The application belongs to the technical field of traffic, and particularly relates to a signal identification method and device of an indicator lamp, vehicle-mounted equipment and a storage medium.

Background

With the continuous development of the technical field of artificial intelligence, unmanned automobiles are widely used, but how to identify the traffic light signals in the current road is a difficult problem in the movement process of the unmanned automobiles.

Currently, when identifying traffic light signals, a pure visual perception scheme is generally adopted. For example, in the process of driving a vehicle to an intersection, if a traffic light is detected from an image captured by an image capturing apparatus, the image is subjected to signal recognition. In addition, in order to improve the recognition accuracy, the photographed image needs to be continuously recognized in the process that the vehicle drives to the intersection, and then a final recognition result is obtained.

However, the method for identifying the traffic light signal has the advantages of general accuracy and high calculation resource consumption.

Disclosure of Invention

The embodiment of the application provides a signal identification method and device of an indicator lamp, vehicle-mounted equipment and a storage medium, which can solve the problems that the accuracy rate of identifying traffic lights at intersections is low and more computing resources are required to be consumed.

In a first aspect, an embodiment of the present application provides a signal identifying method of an indicator light, where the method includes:

Acquiring left-eye images and right-eye images of target indicator lamps at the intersection;

determining a first distance between the vehicle and the target indicator lamp according to the left eye image and the right eye image;

If the first distance is greater than the preset distance, repeating the steps of acquiring left-eye images and right-eye images of the target indicator lights at the intersection and obtaining the first distance;

And if the first distance is smaller than or equal to the preset distance, identifying the target indicator lamp according to the left-eye image or the right-eye image to obtain a traffic signal of the target indicator lamp.

In a second aspect, an embodiment of the present application provides a signal recognition device for an indicator light, including:

the first acquisition module is used for acquiring left-eye images and right-eye images of the target indicator lamps at the intersection;

The first determining module is used for determining a first distance between the vehicle and the target indicator lamp according to the left eye image and the right eye image;

The execution module is used for repeatedly executing the steps of acquiring left-eye images and right-eye images of the target indicator lamps at the intersection and obtaining the first distance if the first distance is greater than the preset distance;

and the identification module is used for identifying the target indicator lamp according to the left eye image or the right eye image if the first distance is smaller than or equal to the preset distance, so as to obtain the traffic signal of the target indicator lamp.

In a third aspect, an embodiment of the present application provides an in-vehicle apparatus, including a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor implementing a method according to the first aspect as described above when executing the computer program.

In a fourth aspect, embodiments of the present application provide a computer readable storage medium storing a computer program which when executed by a processor performs a method as in the first aspect.

In a fifth aspect, an embodiment of the present application provides a computer program product for causing an in-vehicle device to perform the method of the first aspect described above when the computer program product is run on the in-vehicle device.

Compared with the prior art, the embodiment of the application has the beneficial effects that: after the left eye image and the right eye image of the target indicator lamp at the intersection are obtained, the first distance between the vehicle and the target indicator lamp can be determined according to the left eye image and the right eye image. And when the first distance is smaller than or equal to the preset distance, the target indicator lamp is identified according to the left eye image or the right eye image, so that a traffic signal of the target indicator lamp is obtained. Based on the method, in the process that the vehicle drives to the intersection, the target indicator lamps in the left-eye image and the right-eye image which are acquired each time are not required to be identified, so that the required calculation resources are reduced. In addition, when the first distance between the first vehicle and the target indicator lamp is determined, auxiliary determination is not needed by using a high-precision map, so that the cost required for identifying the target indicator lamp is further reduced.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments or the description of the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a flowchart of an implementation of a signal recognition method of an indicator lamp according to an embodiment of the present application;

FIG. 2 is a schematic diagram of an implementation manner of determining a first distance in a signal recognition method of an indicator according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of a signal recognition device of an indicator light according to an embodiment of the present application;

Fig. 4 is a schematic structural diagram of an in-vehicle apparatus according to an embodiment of the present application.

Detailed Description

In the following description, for purposes of explanation and not limitation, specific details are set forth such as the particular system architecture, techniques, etc., in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

Furthermore, the terms "first," "second," "third," and the like in the description of the present specification and in the appended claims, are used for distinguishing between descriptions and not necessarily for indicating or implying a relative importance.

Traffic light identification is the most basic and key important link in unmanned, and a pure visual perception scheme is widely used at present, and can identify traffic lights in a two-dimensional image space, but lacks real-world three-dimensional information (distance between the traffic lights and vehicles), so that false detection and resource waste calculation are easy to cause.

Specifically, when the traffic light is identified in the process that the vehicle drives to the intersection, if the traffic light is detected from the image shot by the camera equipment, the image is processed by adopting a preset identification network so as to determine the signal of the traffic light. In addition, in order to improve the recognition accuracy, it is necessary to continuously recognize the photographed image while the vehicle is driving to the intersection.

However, the traffic light signals identified by the method have a general effect and require more computing resources.

Based on the above, in order to reduce the required computing resources, the embodiment of the application provides a signal recognition method of an indicator lamp, which can be applied to vehicle-mounted equipment in a vehicle. A signal recognition program may be deployed in the in-vehicle apparatus, and the signal recognition program may execute the method in the present embodiment when the vehicle passes through the intersection.

Referring to fig. 1, fig. 1 shows a flowchart of an implementation of a signal recognition method of an indicator lamp according to an embodiment of the present application, where the method includes the following steps:

S101, obtaining left-eye images and right-eye images of target indicator lamps at the intersection.

In an embodiment, the target indicator is an indicator corresponding to an intersection or a lane where the vehicle is traveling. A sensing device, such as an image pickup device and a radar device, is generally provided on a vehicle to sense an external environment. In this embodiment, the vehicle may sense the external environment in real time using the image pickup apparatus.

Specifically, the image capturing apparatus may be a binocular camera, which may acquire image information of the target indicator light through the left and right cameras, respectively. I.e. left and right eye images are acquired. In other embodiments, the image capturing apparatus may also be two monocular cameras, which may acquire left-eye images and right-eye images of the target indicator light, respectively. At this time, when the subsequent steps are performed, the camera internal parameters and external parameters corresponding to the two monocular cameras respectively need to be calibrated in advance.

For example, one or more parameters of focal length, pixel, position and rotation direction of the camera in the world coordinate system of each monocular camera need to be labeled in advance.

The image pickup device can send the collected left-eye images and right-eye images to the vehicle-mounted device in real time or at intervals of preset time so as to be acquired by the vehicle-mounted device.

S102, determining a first distance between the vehicle and the target indicator lamp according to the left eye image and the right eye image.

In an embodiment, the vehicle-mounted device may detect the left-eye image and the right-eye image according to a preset target detection network, so as to determine the area where the target indicator lamp is located in the left-eye image and the right-eye image, and then perform parallax estimation on the area to obtain the distance between the vehicle and the traffic light.

Specifically, the vehicle-mounted device may determine the first distance according to S201 to S203 shown in fig. 2, which is described in detail as follows:

S201, determining a left detection frame containing a target indicator lamp in the left eye image.

In an embodiment, the vehicle-mounted device may determine, according to a preset target detection network, a region where the target indicator lamp is located from the left eye image, and determine the region as a left detection frame.

The target detection network may be yolo detection network, or convolutional neural network, which is not limited thereto. In this embodiment, the vehicle-mounted device may specifically determine a left detection frame including the target indicator light by using yolo detection network.

S202, respectively determining second pixel points corresponding to the first pixel points in the left detection frame one by one from the right eye image.

In one embodiment, the target indicator is also included in the right-eye image because the binocular cameras are capturing the target indicator. Based on this, for any first pixel point in the left detection frame, a corresponding second pixel point should be matched in the right eye image.

Specifically, the vehicle-mounted device may determine, according to a fast nearest neighbor search packet algorithm (Fast Library for Approximate Nearest Neighbors, FLANN) or a feature-based matching algorithm, a second pixel point corresponding to each first pixel point from the right-eye image.

In another embodiment, to improve accuracy of the first distance calculated later, the vehicle-mounted device may also determine a right detection frame including the target indicator from the right-eye image first when determining the corresponding second pixel point from the right-eye image. And then, if the second pixel point corresponding to the first pixel point is determined not to be in the right detection frame, the vehicle-mounted equipment can delete the first pixel point and the second pixel point. That is, the first pixel point processed by the step is located in the left detection frame, and the second pixel point corresponding to the first pixel point is also located in the right detection frame. Further, accuracy in determining the first distance according to all corresponding first pixel points and second pixel points is improved.

S203, determining a first distance according to all the corresponding first pixel points and second pixel points.

In an embodiment, after determining the second pixel points corresponding to all the first pixel points, the vehicle-mounted device may perform parallax estimation on all the first pixel points and the second pixel points to determine the first distance. The parallax in this embodiment means: a certain point in the three-dimensional scene (a target indicator light in the three-dimensional scene) corresponds to the pixel distance of the imaging point in the left-eye image and the right-eye image.

Specifically, for any one group of corresponding first pixel points and second pixel points, the vehicle-mounted device may determine parallax between the first pixel points and the second pixel points. Thereafter, a first distance is determined from all disparities.

For example, the in-vehicle device may determine a first pixel location of a first pixel point on the left-eye image and a second pixel location of a second pixel point on the right-eye image. Then, a distance between the first pixel position and the second pixel position is determined as parallax.

Wherein, when determining the first pixel position and the second pixel position, the pixel coordinate system of the left eye image and the pixel coordinate system of the right eye image should be consistent. Specifically, the vehicle-mounted device may construct a pixel coordinate system of the left-eye image by using a center position or positions of four corner points in the left-eye image as an origin of coordinates. Likewise, for a pixel coordinate system of the right eye image, its origin of coordinates should coincide with the origin of coordinates in the left eye image. For example, the center positions in the left-eye image and the right-eye image are each set as the origin of coordinates.

Wherein, determining the parallax between the first pixel point and the second pixel point may be: the distance between the first pixel position and the second pixel position is determined as parallax. Specifically, after the first pixel position and the second pixel position are obtained, calculation can be performed according to an existing distance formula between two points.

For example, when the first pixel position of the first pixel point is (x, y), and the second pixel position of the corresponding second pixel point is (x-d, y), the parallax is d.

It can be understood that the number of the first pixels in the left detection frame is usually plural, and thus, after the processing based on the above steps, the parallax between the plural first pixels and the corresponding second pixels will be correspondingly obtained. Then, for any parallax, the in-vehicle apparatus may determine a second distance of the vehicle from the target indicator lamp according to the parallax. And then, determining the average value of all the second distances as the first distance.

In an embodiment, the vehicle-mounted device may process the parallax by using a preset camera focal length of the binocular camera and a left-right eye camera optical center distance based on a binocular stereoscopic vision principle, so as to obtain the second distance. Specifically, the vehicle-mounted device may guide the parallax, the focal length of the camera, and the distance between the optical centers of the left and right eye cameras into the following formula to calculate the second distance. The details are as follows:

D＝f*B/d；

Wherein D is the second distance, f is the camera focal length of the image pickup device, and B is the distance between the optical centers of the left and right cameras of the binocular camera. Wherein, when the image pickup apparatus is two monocular cameras, B is the optical center distance between the two monocular cameras.

In one embodiment, a plurality of second distances will be obtained after each parallax is processed by the above formula. In the embodiment of the application, the average value of the plurality of second distances can be determined as the first distance.

As can be seen from the explanation of the steps S201 to S203, when determining the first distance between the image capturing device and the traffic light, the left target image and the right target image of the target indicator light can be acquired by the binocular camera, and the parallax between the first pixel point and the second pixel point is calculated according to the binocular stereoscopic vision principle, so as to determine the first distance according to all the parallaxes. Based on the method, the vehicle-mounted equipment can determine the distance between the vehicle and the traffic light without depending on a high-precision map and distance measuring equipment, so that the use cost is reduced.

In another embodiment, the vehicle-mounted device may further determine a right detection frame including the target indicator light in the right eye image. And then, respectively determining first pixel points corresponding to each second pixel point in the right detection frame from the left eye image, and determining a first distance according to all the corresponding first pixel points and second pixel points. Specific reference is made to the descriptions of S201 to S203 above, and this will not be explained.

And S103, if the first distance is greater than the preset distance, repeating the step of acquiring the left eye image and the right eye image of the target indicator lamp at the intersection and obtaining the first distance.

In an embodiment, the preset distance may be set according to practical situations, which is not limited. For example, the preset distance may be 100m.

In this embodiment, after the first distance is determined, the vehicle-mounted device does not immediately identify the signal of the target indicator lamp in the left-eye image or the right-eye image by using the preset target detection network. It will be appreciated that when the first distance is greater than the predetermined distance, the distance between the vehicle and the target indicator may be considered to be greater. At this time, the target detection network may detect the target indicator from the left-eye image or the right-eye image captured by the binocular camera, but the distance may be relatively long, so that the sharpness of the target indicator in the left-eye image or the right-eye image is relatively low. Further, the accuracy of the in-vehicle apparatus in recognizing the target indicator lamp in the left-eye image or the right-eye image is made low.

Based on this, when it is determined that the first distance is greater than the preset distance, the vehicle-mounted device may acquire a left-eye image and a right-eye image of the target indicator light included in the next frame shot by the binocular camera to repeatedly execute the signal recognition method of the indicator light.

When the target indicator in the left-eye image or the right-eye image of the current frame is not subjected to the subsequent recognition processing, the false detection rate of recognizing the target indicator due to the long distance can be reduced, and the calculation resources required for recognizing the target indicator by the vehicle-mounted equipment can be reduced.

And S104, if the first distance is smaller than or equal to the preset distance, identifying the target indicator lamp according to the left-eye image or the right-eye image, and obtaining the traffic signal of the target indicator lamp.

In an embodiment, when the first distance is smaller than or equal to the preset distance, it may be considered that the distance between the vehicle and the target indicator lamp is relatively short, and the sharpness of the target indicator lamp is relatively high in the left-eye image or the right-eye image shot by the binocular camera. Therefore, when the left-eye image or the right-eye image with higher definition is identified, the accuracy of the traffic signal of the target indicator lamp obtained by the method is also generally higher.

In an embodiment, the traffic signal identifying the target indicator may be the lighting color and the lighting duration of the target indicator, or may be the lighting color of only the target indicator, which is not limited.

For example, when it is determined that the lighting color is green and the lighting time period is 10s, it can be considered that the traffic signal of the target indicator lamp is used to indicate that the traffic rule of running the red light is not violated when the vehicle passes through the intersection within 10 s. Or may be used to indicate that the lighting color of the target indicator light will change from green to red or yellow after 10 s.

In this embodiment, after the left-eye image and the right-eye image of the target indicator at the intersection are obtained, the first distance between the vehicle and the target indicator may be determined according to the left-eye image and the right-eye image. And when the first distance is smaller than or equal to the preset distance, the target indicator lamp is identified according to the left eye image or the right eye image, so that a traffic signal of the target indicator lamp is obtained. Based on the method, in the process that the vehicle drives to the intersection, the target indicator lamps in the left-eye image and the right-eye image which are acquired each time are not required to be identified, so that the required calculation resources are reduced. In addition, when the first distance between the first vehicle and the target indicator lamp is determined, auxiliary determination is not needed by using a high-precision map, so that the cost required for identifying the target indicator lamp is further reduced.

In an embodiment, the embodiment described in S101-S104 is a scene with one traffic light (target indicator light) at the intersection. However, when the intersection is a T-junction or an intersection, the number of traffic lights at the intersection is generally greater. At this time, the in-vehicle apparatus detects a plurality of indicator lamps when detecting the indicator lamps in the left-eye image and the right-eye image. Therefore, when the vehicle-mounted device executes the signal recognition method of the indicator lamp, it is necessary to determine the target indicator lamp corresponding to the current intersection from the plurality of indicator lamps.

Specifically, when there are a plurality of indicator lamps at the intersection, the vehicle-mounted device may determine the third distance between the vehicle and each indicator lamp according to the left-eye image and the right-eye image, respectively. Then, determining the indicator lamp corresponding to the minimum value in all the third distances as a target indicator lamp; and determining the minimum value of all the third distances as the first distance.

It will be appreciated that when there are multiple indicator lights at the intersection, multiple indicator lights will be included in both the left eye image and the right eye image of a frame captured by the binocular camera. At this time, when the vehicle-mounted device detects the left-eye image according to the preset target detection network, a left detection frame corresponding to each indicator light is obtained. Then, for the left detection frame corresponding to any indicator lamp, a third distance between the vehicle and the indicator lamp may be determined according to the steps in S202-S203. And finally, the vehicle-mounted equipment can directly determine the minimum value of the third distances as a target indicator lamp corresponding to the current intersection, and determine the minimum value of the third distances as the first distance.

It should be noted that, after determining the target indicator and the first distance, the vehicle-mounted device may execute the steps in S103 or S104, and when determining that the first distance is greater than the preset distance, repeatedly execute the steps in S101 to S104 on the left eye image or the right eye image of the next frame shot by the binocular camera until the traffic signal of the target indicator is obtained.

It should be added that, after determining the target indicator, if the steps S101-S104 are required to be repeatedly executed, the vehicle-mounted device may directly determine the target indicator from the multiple indicator according to a preset tracking algorithm. Therefore, the vehicle-mounted device can directly acquire the left-eye image and the right-eye image containing the target indicator lamp at the intersection, without acquiring the left-eye image and the right-eye image containing all the indicator lamps again, and execute the step of determining the target indicator lamp from the plurality of indicator lamps again. Furthermore, the computing resources required for the in-vehicle apparatus can be reduced.

In this embodiment, the vehicle-mounted device may determine a third distance between the vehicle and each indicator light by using the binocular stereoscopic vision principle, and determine the target indicator light corresponding to the current intersection according to the plurality of third distances. Therefore, the false detection rate of identifying the traffic signal of the target indicator lamp can be reduced under the traffic scene with a plurality of traffic lights at the intersection. In addition, in the embodiment, when the target indicator lamp is determined, a high-precision map or an additional distance measuring device is not needed, so that the use cost of the vehicle-mounted device is reduced.

Referring to fig. 3, fig. 3 is a block diagram illustrating a signal recognition device of an indicator according to an embodiment of the application. The signal recognition device of the indicator lamp in this embodiment includes modules for performing the steps in the embodiments corresponding to fig. 1 and 2. Refer specifically to fig. 1 and fig. 2 and the related description in the embodiments corresponding to fig. 1 and fig. 2. For convenience of explanation, only the portions related to the present embodiment are shown. Referring to fig. 3, the signal recognition apparatus 300 of the indication lamp may include: a first acquisition module 310, a first determination module 320, an execution module 330, and an identification module 340, wherein:

The first obtaining module 310 is configured to obtain a left eye image and a right eye image of a target indicator at an intersection.

The first determining module 320 is configured to determine a first distance between the vehicle and the target indicator according to the left eye image and the right eye image.

And the execution module 330 is configured to repeatedly execute the step of acquiring the left-eye image and the right-eye image of the target indicator at the intersection and obtaining the first distance if the first distance is greater than the preset distance.

The identifying module 340 is configured to identify the target indicator according to the left-eye image or the right-eye image if the first distance is less than or equal to the preset distance, so as to obtain a traffic signal of the target indicator.

In an embodiment, the first determining module 320 is further configured to:

Determining a left detection frame containing a target indicator lamp in the left eye image; respectively determining second pixel points corresponding to each first pixel point in the left detection frame from the right eye image; and determining a first distance according to all the corresponding first pixel points and the second pixel points.

In an embodiment, the first determining module 320 is further configured to:

for any group of corresponding first pixel points and second pixel points, determining parallax between the first pixel points and the second pixel points; a first distance is determined from all disparities.

In an embodiment, the first determining module 320 is further configured to:

Determining a first pixel position of a first pixel point on the left-eye image; determining a second pixel position of a second pixel point on the right-eye image; the pixel coordinate system of the left eye image is consistent with the pixel coordinate system of the right eye image; the distance between the first pixel position and the second pixel position is determined as parallax.

In an embodiment, the first determining module 320 is further configured to:

determining a second distance between the vehicle and the target indicator lamp according to the parallax for any parallax; and determining the average value of all the second distances as the first distance.

In an embodiment, the first obtaining module 310 is further configured to:

And acquiring a left eye image and a right eye image of the target indicator lamp at the intersection by adopting a binocular camera.

In an embodiment, the first determining module 320 is further configured to:

And processing the parallax according to the camera focal length of the binocular camera and the optical center distance of the left and right eye cameras to obtain a second distance.

In one embodiment, the signal recognition device 300 of the indicator light includes:

and the second determining module is used for determining a third distance between the vehicle and each indicator lamp according to the left eye image and the right eye image if the indicator lamps at the intersection are multiple.

The third determining module is used for determining the indicator lamp corresponding to the minimum value in all the third distances as a target indicator lamp; and determining the minimum value of all the third distances as the first distance.

It should be understood that, in the block diagram of the signal identifying apparatus of the indicator light shown in fig. 3, each module is configured to perform each step in the embodiment corresponding to fig. 1 and 2, and each step in the embodiment corresponding to fig. 1 and 2 is explained in detail in the foregoing embodiment, and specific reference is made to fig. 1 and 2 and related descriptions in the embodiment corresponding to fig. 1 and 2, which are not repeated herein.

Fig. 4 is a block diagram of an in-vehicle apparatus according to an embodiment of the present application. As shown in fig. 4, the in-vehicle apparatus 400 of this embodiment includes: a processor 410, a memory 420, and a computer program 430 stored in the memory 420 and executable on the processor 410, such as a program for a signal recognition method of an indicator light. The processor 410, when executing the computer program 430, implements the steps of the signal identifying method embodiments of the respective indicator lamps described above, such as S101 to S104 shown in fig. 1. Or the processor 410 may perform the functions of the modules in the embodiment corresponding to fig. 3, for example, the functions of the modules 210 to 240 shown in fig. 3, when executing the computer program 430, refer to the related descriptions in the embodiment corresponding to fig. 3.

For example, the computer program 430 may be divided into one or more modules, and one or more modules are stored in the memory 420 and executed by the processor 410 to implement the signal recognition method of the indicator light provided by the embodiment of the present application. One or more of the modules may be a series of computer program instruction segments capable of performing particular functions for describing the execution of the computer program 430 in the in-vehicle apparatus 400. For example, the computer program 430 may implement the signal recognition method of the indicator lamp provided in the embodiment of the present application.

The in-vehicle apparatus 400 may include, but is not limited to, a processor 410, a memory 420. It will be appreciated by those skilled in the art that fig. 4 is merely an example of the in-vehicle device 400 and does not constitute a limitation of the in-vehicle device 400, and may include more or less components than illustrated, or may combine certain components, or different components, e.g., the in-vehicle device may further include an input-output device, a network access device, a bus, etc.

The processor 410 may be a central processing unit, but may also be other general purpose processors, digital signal processors, application specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 420 may be an internal storage unit of the in-vehicle apparatus 400, such as a hard disk or a memory of the in-vehicle apparatus 400. The memory 420 may also be an external storage device of the in-vehicle device 400, such as a plug-in hard disk, a smart memory card, a flash memory card, or the like, provided on the in-vehicle device 400. Further, the memory 420 may also include both an internal storage unit and an external storage device of the in-vehicle device 400.

The embodiment of the application provides a computer readable storage medium, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the signal identification method of the indicator lamp in each embodiment.

Embodiments of the present application provide a computer program product that, when executed on an in-vehicle apparatus, causes the in-vehicle apparatus to execute the signal recognition method of the indicator lamp in each of the above embodiments.

The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application.

Claims (10)

1. A method of signal identification for an indicator light, the method comprising:

Acquiring left-eye images and right-eye images of target indicator lamps at the intersection;

determining a first distance between a vehicle and the target indicator lamp according to the left eye image and the right eye image;

If the first distance is greater than the preset distance, repeating the steps of acquiring left-eye images and right-eye images of the target indicator lights at the intersection and obtaining the first distance;

And if the first distance is smaller than or equal to the preset distance, identifying the target indicator lamp according to the left eye image or the right eye image to obtain a traffic signal of the target indicator lamp.

2. The method of claim 1, wherein the determining a first distance of a vehicle from the target indicator from the left eye image and the right eye image comprises:

determining a left detection frame containing the target indicator lamp in the left eye image;

Respectively determining second pixel points corresponding to each first pixel point in the left detection frame from the right eye image;

And determining the first distance according to all the corresponding first pixel points and the second pixel points.

3. The method of claim 2, wherein the determining the first distance from all corresponding first pixel points and second pixel points comprises:

Determining parallax between the first pixel point and the second pixel point according to any group of the corresponding first pixel point and the second pixel point;

and determining the first distance according to all the parallaxes.

4. A method according to claim 3, wherein said determining the disparity between the first pixel point and the second pixel point comprises:

Determining a first pixel position of the first pixel point on the left-eye image;

determining a second pixel position of the second pixel point on the right-eye image; the pixel coordinate system of the left eye image is consistent with the pixel coordinate system of the right eye image;

and determining a distance between the first pixel position and the second pixel position as the parallax.

5. A method according to claim 3, wherein said determining said first distance from all of said disparities comprises:

determining a second distance between the vehicle and the target indicator lamp according to the parallax for any parallax;

and determining the average value of all the second distances as the first distance.

6. The method of claim 5, wherein the acquiring left and right eye images of the target indicator light at the intersection comprises:

A binocular camera is adopted to obtain a left eye image and a right eye image of a target indicator lamp at the intersection;

the determining the second distance between the vehicle and the target indicator according to the parallax comprises the following steps:

and processing the parallax according to the camera focal length of the binocular camera and the distance between the optical centers of the left and right eye cameras to obtain the second distance.

7. The method of any of claims 1-6, wherein determining a first distance of a vehicle from the target indicator from the left eye image and the right eye image comprises:

If a plurality of indicator lamps are arranged at the intersection, determining a third distance between the vehicle and each indicator lamp according to the left eye image and the right eye image;

Determining the indicator lamps corresponding to the minimum value in all the third distances as the target indicator lamps; and determining the minimum value of all the third distances as the first distance.

8. A signal recognition device for an indicator light, the device comprising:

the first acquisition module is used for acquiring left-eye images and right-eye images of the target indicator lamps at the intersection;

The first determining module is used for determining a first distance between the vehicle and the target indicator lamp according to the left eye image and the right eye image;

The execution module is used for repeatedly executing the steps of acquiring left-eye images and right-eye images of the target indicator lamps at the intersection and obtaining the first distance if the first distance is larger than the preset distance;

And the identification module is used for identifying the target indicator lamp according to the left eye image or the right eye image if the first distance is smaller than or equal to the preset distance, so as to obtain a traffic signal of the target indicator lamp.

9. An in-vehicle device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the method according to any one of claims 1 to 7.