Early warning method and device for red light running of vehicle and storage medium

Abstract

The invention relates to the technical field of vehicle driving control, in particular to an early warning method and device for red light running of a vehicle and a storage medium. The method comprises the following steps: when a vehicle is in a driving state, acquiring images around the vehicle in real time; identifying the image to obtain a target lane where the vehicle is located, a current signal lamp state of each lane and a current distance of the vehicle relative to a lane stop line; according to the current gear of the vehicle, the driving distance in the future time period, the current distance and the current signal lamp state of the target lane, whether the risk of running the red light exists or not is determined; if the vehicle has the red light running risk, red light running early warning is carried out. The red light running risk of the vehicle in the future time period is predicted, early warning is carried out, and the driving safety can be improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle driving control technology, and in particular to a vehicle running a red light warning method, device and storage medium. Background Art

[0002] The in-vehicle navigation system generally has a navigation broadcast function, such as there is a crossroads ahead, the current light is red, etc., to remind the driver.

[0003] However, due to the limited navigation accuracy of the in-vehicle navigation system and the unstable in-vehicle network signal, the red light at the intersection ahead may not be recognized, and the driver may not be given an accurate warning in time. Moreover, the driver may not turn on the navigation system when he is familiar with the road, and the driver cannot be effectively reminded.

[0004] It can be seen that if the driver is at risk of running a red light, it is difficult to effectively and promptly remind the driver by relying solely on the existing vehicle navigation system, and it is also impossible to effectively intervene in the moving vehicle in a timely manner. Summary of the invention

[0005] The purpose of this application is to provide a vehicle red light running warning method, device and storage medium, accurately predict the risk of red light running in the future period, and issue a warning, which is conducive to improving driving safety.

[0006] In order to achieve the above objectives, this application adopts the following technical solutions: In a first aspect, the present application provides a vehicle red light running warning method, comprising: When the vehicle is in a driving state, acquiring images around the vehicle in real time; Recognize the image to obtain the target lane where the vehicle is located, the current signal light status of each lane, and the current distance of the vehicle relative to the stop line of the lane; Determining whether there is a risk of running a red light according to the current gear position of the vehicle, the travel distance in the future period, the current distance and the current signal light state of the target lane; If the vehicle has a risk of running a red light, a red light running warning is issued.

[0007] In a second aspect, the present application provides an electronic device, including: at least one processor, and a memory communicatively coupled to at least one of the processors; The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can execute any of the vehicle red light running warning methods.

[0008] In a third aspect, the present application provides a computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions are used to enable the computer to execute any of the above-described vehicle red light running warning methods.

[0009] Compared with the prior art, the beneficial effects of this application are: This application acquires and identifies images around the vehicle in real time, so as to know in real time the target lane where the vehicle is located, the status of the traffic lights in each lane, and the current distance of the vehicle relative to the stop line of the lane; then, based on the current gear of the vehicle, the driving distance in the future period, the current distance, and the current status of the traffic lights, it determines whether there is a risk of running a red light. It comprehensively considers the driving / parking status of the vehicle, the current position of the vehicle, the future position of the vehicle, and the current status of the traffic lights, and can accurately assess whether the vehicle has a risk of running a red light, with high fault tolerance; further, if there is a risk of running a red light, a red light running warning is issued. This application uses image recognition results combined with the vehicle's gear position and predicted future driving distance to assess whether there is a risk of running a red light, and automatically triggers a warning. It does not rely on the navigation system for red light running risk identification and warning, nor does it rely on other hardware, and has low implementation cost and high reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0011] Figure 1 It is a flow chart of a vehicle red light running warning method provided in an embodiment of the present application; Figure 2 is a schematic diagram of a vehicle driving scenario provided by an embodiment of the present invention; Figure 3 It is a flow chart of another vehicle red light running warning method provided in an embodiment of the present application; Figure 4 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] The following is a description of exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0013] Figure 1 This is a flow chart of a vehicle red light running warning method provided by this embodiment. This embodiment is applicable to the scenario of predicting whether a vehicle has a risk of running a red light in the future and issuing a warning. The method can be executed by an electronic device. The electronic device can be integrated in the vehicle computer, or it can be independent of the vehicle computer and communicated with the vehicle computer.

[0014] Figure 2 Schematic diagram of a vehicle driving scene provided by an embodiment of the present invention. The vehicle is driving on a road having at least one lane. Figure 2 Three lanes are shown, with arrows drawn on each lane to indicate a left turn lane, a straight lane, and a right turn lane. Of course, multiple lanes such as a U-turn lane, a straight and left turn lane may also be provided. Figure 2 Only some lane types are drawn. A lane stop line is drawn at the end of each lane, and there is a corresponding signal light. It appears red, yellow or green. Figure 2 In the example, a vehicle is traveling in a straight lane, and the traffic light in the straight lane is red. If it is predicted that the vehicle will go out of the stop line of the lane, there is a risk of running a red light.

[0015] like Figure 1 As shown, this embodiment provides a warning method for vehicles running red lights, comprising the following steps: S110: When the vehicle is in a driving state, acquire images around the vehicle in real time.

[0016] The vehicle is in the driving state (Driving state), including the state of vehicle speed greater than 0 and the state of vehicle about to drive. The following conditions can be used to determine whether the vehicle is about to drive: 1. The vehicle speed is equal to 0; 2. The gear is in the forward gear (D gear); 3. The main driver is detected; 4. The vehicle is in a high-voltage state; 5. The vehicle user mode is in the driving mode. When the above five conditions are met at the same time, it is confirmed that the vehicle is in the state of about to drive.

[0017] Whether the vehicle is driving or about to drive, it is necessary to detect whether there is a risk of running a red light. Therefore, the vehicle is in driving state as a trigger condition, and the image around the vehicle is acquired in real time through the on-board camera. If the vehicle is not in driving state, that is, InDriving state, the vehicle speed, gear position, main driver's seat status, high voltage status and user mode will continue to be monitored.

[0018] Optionally, there are multiple on-board cameras, which can capture images or videos directly in front of the vehicle, directly above, in front of the left, in front of the right, in the upper left and in the upper right. Note that the "left" and "right" here are relative to the driver's perspective. If a video is captured, the video frames are sampled at a set frequency to obtain images at the corresponding position. Of course, if the camera has a wide angle, a wider field of view can be captured, or one camera can be used to capture the image in front of the vehicle.

[0019] S120: Recognize the image to obtain the target lane where the vehicle is located, the current signal light status of each lane, and the current distance of the vehicle relative to the stop line of the lane.

[0020] After each image is acquired, the image is recognized. Optionally, an image recognition algorithm is used to recognize the lanes in the image, the current signal light status of each lane, and the current distance of the vehicle relative to the lane stop line.

[0021] by Figure 2 For example, the lanes include left-turn lanes, through lanes, and right-turn lanes. The target lane where the vehicle is located is the through lane. The current signal light status of the left-turn lane is green, the current signal light status of the through lane is red, and the current signal light status of the right-turn lane is green. The current distance of the vehicle relative to the lane stop line can be the horizontal distance between the front of the vehicle and the lane stop line, for example, 5m.

[0022] S130: Determine whether there is a risk of running a red light based on the current gear position of the vehicle, the driving distance in the future period, the current distance, and the current traffic light status of the target lane.

[0023] The gears of a vehicle include forward gear (D gear), parking gear (P gear) and neutral gear (N gear). If the vehicle is in D gear, it means that the vehicle is moving and there may be a risk of running a red light. If the vehicle is in P gear or N gear, it means that the vehicle is parked or temporarily stopped and there is no risk of running a red light.

[0024] The future period is the duration of at least one image sampling cycle in the future starting from the current moment. The driving distance of the vehicle in the future period refers to the length of the route that the vehicle will travel in the future period, which can be predicted by the kinematic formula.

[0025] Optionally, the vehicle's driving distance in a future time period is subtracted from the current distance to obtain the relative position of the vehicle to the lane stop line in the future time period; if the vehicle's current gear is a forward gear, the current signal light status of the target lane is a yellow light or a red light, and the vehicle crosses the lane stop line in the future time period, it is determined that the vehicle has a risk of running a red light.

[0026] For example, if the vehicle is in D gear, and the current distance from the lane stop line is greater than or equal to the vehicle's travel distance in the future period, it means that the vehicle may cross the lane stop line in the future period, and when the current signal light status of the target lane is red or yellow, there is a risk of running a red light. If the vehicle is in P gear or N gear, or the current distance from the lane stop line is less than the vehicle's travel distance in the future period, or the current signal light status of the target lane is green, it means that the vehicle will not cross the lane stop line in the future period, and there is no risk of running a red light.

[0027] S140: If the vehicle has a risk of running a red light, issue a red light running warning.

[0028] The purpose of the red light running warning is to effectively and timely remind the driver that he is about to run a red light, so that the driver can take braking measures to avoid the occurrence of red light running. Based on this, this embodiment controls the components in the vehicle to provide reminder information to the driver. For example, a "red light running" voice is issued through the vehicle speaker or the large screen of the vehicle computer; the red light running icon is displayed / flashed through the display device such as the large screen of the vehicle computer, the instrument panel, or the head-up display; the main driver's seat is controlled to vibrate or the steering wheel is controlled to vibrate to remind the driver that there is a risk of running a red light.

[0029] When the red light running intervention function of the vehicle is turned on, the vehicle is controlled to slow down and / or stop. In some cases, the in-vehicle components are controlled to provide a reminder message to the driver and then start timing. When the set time (such as the driver's reaction time) is reached, if the driver does not brake the vehicle, it is determined whether the red light running intervention function of the vehicle is turned on. If it is turned on, a braking request is sent to the vehicle controller to control the vehicle to slow down and / or stop.

[0030] Optionally, if the vehicle does not have the risk of running a red light, the process returns to S110 to continue acquiring images around the vehicle and perform subsequent operations to monitor in real time whether the vehicle has the risk of running a red light.

[0031] Compared with the prior art, the beneficial effects of the embodiments of the present application are: This application acquires and identifies images around the vehicle in real time, so as to know in real time the target lane where the vehicle is located, the status of the traffic lights in each lane, and the current distance of the vehicle relative to the stop line of the lane; then, based on the vehicle's current gear, the distance traveled in the future period, the current distance, and the current status of the traffic lights, it determines whether there is a risk of running a red light. It comprehensively considers the vehicle's driving / parking status, the vehicle's current position, future position, and current status of the traffic lights, and can accurately assess whether the vehicle has a risk of running a red light; further, if there is a risk of running a red light, a red light running warning is issued. This application uses image recognition results combined with the vehicle's gear position and predicted future driving distance to assess whether there is a risk of running a red light, and automatically triggers a warning. It does not rely on the navigation system for red light running risk identification and warning, nor does it rely on other hardware, and has low implementation cost and high reliability.

[0032] Figure 3 is a flow chart of another vehicle red light running warning method provided by an embodiment of the present application. This embodiment refines the image recognition process based on the above embodiment. Figure 3 , the method provided in this embodiment includes the following operations: S210: When the vehicle is in a driving state, acquire images around the vehicle in real time.

[0033] Multiple cameras are used to collect images from different perspectives around the vehicle in real time. Multiple cameras use unified timing and timestamp each captured image.

[0034] S220 , pre-processing the images taken at the same time and at various viewing angles around the vehicle to obtain a plurality of processed images.

[0035] Filter multiple images taken at the same time based on the timestamp on the image. For example, image 1, image 2, image 3, and image 4 taken at time t0. The original image taken by the camera has a lot of noise and needs to be preprocessed. The preprocessing operation includes but is not limited to illumination correction, filtering and denoising, and edge enhancement. The image after preprocessing is called the processed image.

[0036] S230: stitching the multiple processed images to obtain a panoramic image.

[0037] Since the multiple processed images are taken from different perspectives, they need to be stitched into a panoramic image to create a wider field of view than a single image, so that the current traffic light status of all lanes, the target lane where the vehicle is located, and the current distance of the vehicle relative to the lane stop line can be obtained through one recognition.

[0038] Multi-view image stitching aims to seamlessly merge images from multiple different viewpoints into a panoramic image. The stitching process includes: 1. Image preprocessing: Perform necessary preprocessing on each image, such as adjusting the size, contrast and brightness, to ensure the consistency of stitching. 2. Feature point detection: Use a specific algorithm (such as Harris corner detection) to extract feature points (such as corners, edges, etc.) from each image. These feature points will serve as the basis for subsequent image registration and stitching. 3. Feature point description and matching: Describe the image area around each feature point and generate feature descriptors. Then, by comparing the similarity between feature descriptors, find the matching feature point pairs between different images. 4. Image registration: After finding the matching feature point pairs, image transformations (such as affine transformations, homography transformations, etc.) are needed to align the images from different viewpoints to the same coordinate system. 5. Image stitching and fusion: Stitch the aligned images, and use image fusion techniques (such as weighted averaging, multi-resolution fusion, etc.) to smooth the stitching seams so that the stitched images can achieve a seamless fusion effect visually. 6. Post-processing: Perform post-processing operations on the stitched panoramic image, such as removing stitching marks, image trimming, etc., to further improve image quality.

[0039] S240: Input the panoramic image into a pre-trained neural network model to obtain the target lane where the vehicle is located, the current traffic light status of each lane, and the current distance of the vehicle relative to the stop line of the lane.

[0040] The neural network model includes at least an input layer, multiple convolutional layers, and an output layer. Optionally, considering that the neural network model is ultimately applied to the vehicle terminal, limited by the terminal computing power and storage capacity, the lightweight convolutional network Mobile Next X can be used as the backbone network. Of course, as the computing power of the terminal increases, other types of neural networks can also be used, such as the YOLO model and the U-Net model.

[0041] The neural network model is trained using training samples. The training samples are multiple panoramic images collected in advance, each of which is marked with the target lane where the vehicle is located, the current signal light status of each lane, and the current distance of the vehicle relative to the lane stop line. The training samples are divided according to a set ratio, such as 8:2, with one part used for model training and the other part used for model testing.

[0042] During training, a panoramic image is represented as a feature vector , i represents the image number, and the formula is as follows: ; in, represents the width of the i-th panoramic image, represents the length of the i-th panoramic image, C represents the number of channels of the panoramic image, and for RGB images, the number of channels is 3. The output of the neural network model is Y i , there is the following formula: ; in, The elements in the sequence from left to right represent: the current signal light status of the left-turn lane, the current signal light status of the through lane, the current signal light status of the right-turn lane, the current signal light status of the U-turn lane, the target lane where the vehicle is located, and the current distance of the vehicle from the stop line of the lane. , , , .exist In the figure, 0 is green light, 2 is red light, and 1 is yellow light. , among which 1 is the left turn lane, 2 is the through lane, 3 is the right turn lane, and 4 is the U-turn lane. , 0 means the current distance is less than 1m, 3 means the current distance is between 1m and 3m, 5 means the current distance is between 3m and 5m, 10 means the current distance is between 5m and 10m, 15 means the current distance is between 10m and 15m, and 20 means the current distance is greater than 15m.

[0043] For each training sample , where the eigenvector , output the result The training batch was set to 10 times and the model iteration number was set to 50 times. The cross entropy loss function was used to evaluate the difference between the model prediction value and the true value. The hidden layer parameters of the neural network model were updated using the stochastic gradient descent algorithm so that the difference between the prediction value of the neural network model and the true value was less than the set threshold.

[0044] After the model training is completed, the neural network model can be evaluated using evaluation indicators such as accuracy, precision, recall, and F1-score. The model that passes the evaluation will be used in the subsequent prediction process.

[0045] After the model training is completed, the stitched panoramic image is input into the neural network model to obtain the target lane where the vehicle is located (i.e. The current traffic light status of each lane (i.e. The value of the signal light indicates the state of the traffic light) and the current distance of the vehicle relative to the stop line of the lane (i.e. The value of represents the current distance).

[0046] S250: Determine whether there is a risk of running a red light according to the current gear position of the vehicle, the travel distance in the future period, the current distance, and the current signal light state of the target lane. If there is no risk of running a red light, return to S210.

[0047] In an optional implementation, first, the current gear position of the vehicle is quantified as a first value, the current traffic light state of the target lane is quantified as a second value, and the vehicle's driving distance in the future period is subtracted from the current distance to obtain a third value.

[0048] Assume that the D gear is quantized to 1, and the P gear and N gear are quantized to 0, which is recorded as G. In the current signal light state, 0 is a green light, 1 is a red light, and 2 is a yellow light. The first value, the second value, and the third value are mathematically operated to obtain the red light running risk value T. For example, the first value, the second value, and the third value are multiplied to obtain the red light running risk value T, as shown in the following formula: ; in, The quantized value representing the current signal light state of the target lane, i.e., the second value. G is the first value, and S is the driving distance of the vehicle in the future period. If the red light running risk value T exceeds the set range, it is determined that the vehicle has a red light running risk, see the following formula:

[0049] If T>0, E=1, there is a risk of running a red light. , E=0, no risk of running a red light.

[0050] S260: If the vehicle is at risk of running a red light, issue a red light running warning.

[0051] Optionally, a warning strategy is determined according to the size of the red light running risk value; and a red light running warning is performed according to the warning strategy.

[0052] exist In the figure, 2 is a red light and 1 is a yellow light. If the current signal light is a yellow light, the driver may pass through the intersection during the yellow light period, and the risk of running a red light will be lower than when the current signal light is a red light. Therefore, referring to the calculation formula of T, if the current signal light is a red light, the calculated risk value of running a red light is greater than the case where the current signal light is a yellow light. Therefore, in order to enable the driver to pass the intersection smoothly during the yellow light period, the components in the vehicle can be controlled to provide reminder information to the driver without slowing down and / or braking. With the real-time acquisition of images around the vehicle, and the identification of the status of the signal lights therein. If a red light is identified, the components in the vehicle continue to be controlled to provide reminder information to the driver; at the same time, when the red light running intervention enable of the vehicle is turned on, the vehicle is controlled to slow down and / or brake to a stop.

[0053] In the above embodiment and the following embodiment, before determining whether there is a risk of running a red light based on the current gear of the vehicle, the driving distance in the future period, the current distance and the current traffic light status of the target lane, it is necessary to predict the driving distance in the future period.

[0054] Optionally, since the camera's sampling period for images is relatively short, the vehicle's driving distance within a sampling period can be predicted, as shown in the following formula: S ; Where V is the current vehicle speed, is the vehicle speed at the previous moment, and t is the duration of a sampling period.

[0055] In some dangerous scenarios, it is necessary to timely predict the driving distance in the future period so as to remind the driver as soon as possible. Therefore, it is possible to extend the sampling period to predict the risk value of running a red light in a longer period of time. For example, the length of the future period is adjusted according to the warning coefficient to obtain a new future length; the driving distance of the vehicle in the future period is calculated according to the current speed of the vehicle, the current acceleration and the new future length. See the following formula: S' ; Among them, ɑ is the warning coefficient. Multiply the warning coefficient by the sampling period t to get the new future duration. S' is the driving distance of the vehicle in the future period.

[0056] The warning coefficient determines the length of the future period, which can be determined based on the current speed of the vehicle and / or the driver's style. The greater the current speed, the greater the warning coefficient, and the more aggressive the driver's style, the greater the warning coefficient, so that it can be predicted whether the vehicle has a risk of running a red light in the more distant future. For example, the current speed is divided by a fixed value to obtain the warning coefficient, or the driver's style is quantified into a fourth value according to the degree of aggressiveness, the more aggressive the fourth value, and the fourth value is used as the warning coefficient. By measuring the acceleration, deceleration and travel time of the vehicle during the following process, the standardized driver aggressiveness index SDAI value is calculated. A larger SDAI value means that the driver adopts a more aggressive following behavior, such as stepping on the accelerator and brake harder. The SDAI value is used as the fourth value. The warning coefficient can also be obtained by dividing the current speed by a fixed value and then multiplying or adding it with the fourth value.

[0057] This embodiment determines the warning coefficient according to the degree of driving danger and extends the sampling period, thereby predicting the risk value of running a red light in a longer time period, giving the driver sufficient time to brake the vehicle and improving driving safety.

[0058] This embodiment provides an electronic device, see Figure 4, includes at least one processor 301 and a memory 302 that is communicatively connected to the at least one processor 301.

[0059] The memory 302 stores instructions that can be executed by at least one of the processors 301. The instructions are executed by at least one of the processors 301 so that at least one of the processors 301 can execute the above-mentioned vehicle red light running warning method, thereby having at least the same advantages as the above-mentioned method.

[0060] Optionally, the electronic device also includes interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on a memory to display graphical information of a GUI (Graphical User Interface) on an external input / output device (such as a display device coupled to an interface). In other embodiments, if necessary, multiple processors can be used together with multiple memories, and / or multiple buses can be used together with multiple memories. Similarly, multiple electronic devices can be connected (for example, as a server array, a group of blade servers, or a multi-processor system), and each device provides some necessary operations.

[0061] The memory 302 is a computer-readable storage medium that can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the vehicle red light running warning method in the embodiment of the present application. The processor 301 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory 301, that is, realizing the above-mentioned vehicle red light running warning method.

[0062] The memory 301 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 302 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 302 may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0063] The electronic device may further include: an input device 303 and an output device 304. The processor 301, the memory 301, the input device 303 and the output device 304 may be connected via a bus or other means.

[0064] The input device 303 can receive input digital or character information, and the output device 304 can include a display device, an auxiliary lighting device (e.g., LED), and a tactile feedback device (e.g., a vibration motor), etc. The display device can include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some embodiments, the display device can be a touch screen.

[0065] This embodiment provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable the computer to execute the above method. The computer instructions on the computer-readable storage medium are used to enable the computer to execute the above method, and thus have at least the same advantages as the above method.

[0066] The medium in the present invention can adopt any combination of one or more computer-readable media. The medium can be a computer-readable signal medium or a computer-readable storage medium. The medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples of the medium (non-exhaustive list) include: 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 this document, the medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, a device or a device or used in combination with it.

[0067] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0068] The program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF (Radio Frequency), etc., or any suitable combination of the above.

[0069] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including 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 user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through 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).

[0070] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this application can be executed in parallel, sequentially or in different orders, as long as the expected results of the technical solution disclosed in this application can be achieved, and this document is not limited here.

[0071] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A vehicle red light running warning method, characterized in that: include: When the vehicle is in a driving state, acquiring images around the vehicle in real time; Recognize the image to obtain the target lane where the vehicle is located, the current signal light status of each lane, and the current distance of the vehicle relative to the stop line of the lane; Determining whether there is a risk of running a red light according to the current gear position of the vehicle, the travel distance in the future period, the current distance and the current signal light state of the target lane; If the vehicle has a risk of running a red light, a red light running warning is issued.

2. The method according to claim 1, characterized in that The image is recognized to obtain the target lane where the vehicle is located, the current signal light status of each lane, and the current distance of the vehicle relative to the lane stop line, including: Preprocessing the images taken at the same time and at various viewing angles around the vehicle to obtain multiple processed images; Stitching the multiple processed images to obtain a panoramic image; The panoramic image is input into a pre-trained neural network model to obtain the target lane where the vehicle is located, the current signal light status of each lane, and the current distance of the vehicle relative to the lane stop line.

3. The method according to claim 1, characterized in that Before determining whether there is a risk of running a red light according to the current gear position of the vehicle, the travel distance in the future period, the current distance and the current signal light state of the target lane, the method further includes: The duration of the future period is adjusted according to the warning coefficient to obtain a new future duration; The driving distance of the vehicle in the future period is calculated according to the current speed of the vehicle, the current acceleration and the new future duration.

4. The method according to claim 3, characterized in that Before adjusting the duration of the future time period according to the warning coefficient to obtain a new future duration, it also includes: The warning coefficient is determined according to the current vehicle speed and / or driver style of the vehicle.

5. The method according to claim 1, characterized in that Determining whether there is a risk of running a red light according to the current gear position of the vehicle, the travel distance in the future period, the current distance and the current signal light state of the target lane includes: Subtracting the travel distance of the vehicle in the future time period from the current distance to obtain the relative position of the vehicle to the lane stop line in the future time period; If the current gear of the vehicle is the forward gear, the current signal light state of the target lane is a yellow light or a red light, and the vehicle crosses the stop line of the lane in a future period, it is determined that the vehicle has a risk of running a red light.

6. The method according to claim 5, characterized in that If the current gear of the vehicle is the forward gear, the current signal light state of the target lane is yellow or red, and the vehicle crosses the stop line of the lane in the future period, then the vehicle is determined to have a risk of running a red light, including: quantizing the current gear position of the vehicle as a first value, quantifying the current signal light state of the target lane as a second value, and subtracting the current distance from the driving distance of the vehicle in the future period to obtain a third value; Performing mathematical operations on the first value, the second value, and the third value to obtain a red light running risk value; If the red light running risk value exceeds a set range, it is determined that the vehicle has a red light running risk.

7. The method according to claim 6, characterized in that If the vehicle has a risk of running a red light, a red light running warning is issued, including: Determine the early warning strategy based on the risk value of running a red light; A red light running warning is performed according to the warning strategy.

8. The method according to any one of claims 1 to 7, characterized in that: If the vehicle has a risk of running a red light, a red light running warning is issued, including: If the vehicle has a risk of running a red light, controlling the components in the vehicle to provide a reminder message to the driver; When the red light running intervention enable of the vehicle is turned on, the vehicle is controlled to decelerate and / or stop.

9. An electronic device, characterized in that: include: at least one processor, and a memory communicatively coupled to at least one of the processors; The memory stores instructions executable by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can execute the vehicle red light running warning method described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The medium stores computer instructions, and the computer instructions are used to enable the computer to execute the vehicle red light running warning method according to any one of claims 1 to 8.

Images