Front vehicle start-up identification method and device, electronic equipment and storage medium

Abstract

Embodiments of the present application provide a front vehicle starting identification method and device, electronic equipment and storage medium. The front vehicle starting identification method comprises: after a current vehicle enters a waiting front vehicle starting state, collecting at least two front images of the current vehicle; obtaining at least two target images based on the at least two front images; based on the at least two target images, determining whether the brake light on the front vehicle of the current vehicle changes from on to off; when the brake light on the front vehicle changes from on to off, determining that the front vehicle starts. In the embodiments of the present application, on the one hand, since the phenomenon that the brake light changes from on to off occurs earlier than the phenomenon that the vehicle moves when the front vehicle starts, the timeliness of the front vehicle starting identification can be improved; on the other hand, the phenomenon that the brake light changes from on to off is not easily affected by other interference factors in the image, so the accuracy of the front vehicle starting identification can be improved.

Description

Technical Field

[0001] This application relates to the field of vehicle networking technology, and in particular to a method, device, electronic device and storage medium for identifying the starting of a preceding vehicle. Background Technology

[0002] With the continuous development of technology and the increasing demands of users, vehicles have become a common choice for travel, bringing great convenience to users. Furthermore, with the rapid development of vehicle networking technology, vehicles are becoming increasingly intelligent.

[0003] While driving, there will be scenarios where a vehicle is stopped and waiting for the vehicle in front to start moving. In such scenarios, the driver often fails to notice the vehicle in front starting moving due to reasons such as a lack of concentration.

[0004] To address the aforementioned issues, existing technologies acquire images of the vehicle's path ahead and only recognize the vehicle's movement when there are significant pixel changes in the image, subsequently providing a voice prompt to the driver. However, this method requires the vehicle ahead to have moved a considerable distance before determining whether it has started moving, and interference from other factors can also lead to misidentification of the vehicle's movement. Therefore, the timeliness and accuracy of recognizing the vehicle's movement ahead are poor. Summary of the Invention

[0005] One objective of the embodiments of this application is to provide a method, device, electronic device, and storage medium for identifying a vehicle starting ahead, which has the advantage of improving the timeliness and accuracy of identifying a vehicle starting ahead based on the phenomenon that the brake lights of the vehicle ahead change from being on to being off.

[0006] Another objective of the embodiments of this application is to provide a method, device, electronic device and storage medium for identifying the start of a preceding vehicle. Its advantage is that it can identify the start of a preceding vehicle by combining the phenomenon of the preceding vehicle moving and the phenomenon of the preceding vehicle's brake lights changing from being lit to being extinguished, thereby improving the efficiency of the identification of the start of a preceding vehicle and reducing the possibility of false identification of the start of a preceding vehicle.

[0007] Another objective of the embodiments of this application is to provide a method, device, electronic device and storage medium for identifying the start of a vehicle in front. Its advantage is that by capturing a portion of the image in front of the vehicle to identify the phenomenon of the brake light changing from on to off, the amount of computation is reduced and the computational efficiency is improved, thereby enhancing the timeliness of the start-up recognition of the vehicle in front.

[0008] Another objective of the embodiments of this application is to provide a method, device, electronic device and storage medium for identifying the start of a vehicle in front. Its advantage is that it identifies the phenomenon of the brake light changing from on to off based on the target channel pixel value of the image in front of the vehicle, which reduces the amount of computation, improves the computational efficiency, and is not easily affected by other interference factors, thereby improving the timeliness and accuracy of the start of the vehicle in front.

[0009] Another objective of the embodiments of this application is to provide a method, device, electronic device and storage medium for identifying the starting of a vehicle in front, which has the advantage of being able to identify the starting of a vehicle in front in the current lane or any lane in an adjacent lane, and further being able to promptly notify the user, which helps to ensure the traffic efficiency of the current vehicle.

[0010] To achieve the above objectives, in a first aspect, a method for recognizing the starting of a preceding vehicle is provided, the method comprising:

[0011] After the current vehicle enters the waiting state for the vehicle in front to start, at least two frontal images of the current vehicle are acquired;

[0012] At least two target images are obtained based on the at least two front images;

[0013] Based on the at least two target images, determine whether the brake lights on the vehicle in front of the current vehicle have changed from being on to being off;

[0014] When the brake lights on the vehicle ahead change from on to off, it is determined that the vehicle ahead has started moving.

[0015] Secondly, a device for recognizing a vehicle starting ahead is provided, the device including a processor and a camera;

[0016] The camera is used to capture at least two frontal images of the current vehicle after the current vehicle enters a waiting state for the vehicle in front to start, and to transmit the at least two frontal images to the processor;

[0017] The processor includes:

[0018] The acquisition module is configured to receive the at least two front images and acquire at least two target images based on the at least two front images;

[0019] The first judgment module is used to determine, based on the at least two target images, whether the brake light on the vehicle in front of the current vehicle changes from lit to off; when the brake light on the vehicle in front changes from lit to off, the module determines that the vehicle in front has started moving.

[0020] Thirdly, an electronic device is provided, comprising: one or more processors; and one or more computer-readable storage media having instructions stored thereon; wherein, when the instructions are executed by the one or more processors, the processors perform the preceding vehicle start-up recognition method as described in any of the preceding claims.

[0021] Fourthly, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, causes the processor to perform the preceding vehicle start-up recognition method as described in any of the preceding claims. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some drawings of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of a vehicle according to an embodiment of this application.

[0024] Figure 2 This is a flowchart illustrating the steps of a method for identifying the starting of a vehicle in front, according to an embodiment of this application.

[0025] Figure 3 This is a flowchart illustrating how to determine the area where the brake light is located, according to an embodiment of this application.

[0026] Figure 4 This is a flowchart of an embodiment of the present application for determining whether the brake light on the vehicle in front of the current vehicle has changed from being lit to being off, based on at least two target images.

[0027] Figure 5 This is a flowchart illustrating how, based on changes in the target channel pixel values ​​of candidate pixels in at least two target images, a process is used to determine whether the brake light on the vehicle in front of the current vehicle has changed from on to off.

[0028] Figure 6 This is a flowchart illustrating the steps of another method for identifying the starting of a vehicle in front, according to an embodiment of this application.

[0029] Figure 7 This is a structural block diagram of a vehicle starting recognition device according to an embodiment of this application.

[0030] Figure 8 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0031] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0032] As mentioned earlier, existing technologies determine whether a vehicle in front has started moving by acquiring an image of the area in front of the vehicle and analyzing the changes in pixels in that image. However, this method requires the vehicle in front to have moved a considerable distance before it can be determined that the vehicle has started moving, resulting in untimely voice prompts. Furthermore, interference from other factors (such as pedestrians crossing in front of the vehicle) can also lead to misidentification of the vehicle's movement, resulting in false voice prompts. Therefore, the timeliness and accuracy of vehicle start-up recognition are both poor.

[0033] The embodiments in this application can be applied to vehicles. See also... Figure 1 The diagram illustrates a vehicle according to an embodiment of this application. Figure 1 As shown, vehicle 10 may include processor 101, camera 102, speaker 103, display 104, and headlights 105. Camera 102 can capture an image of the vehicle's front and transmit it to processor 101. Processor 101 can analyze the image of the vehicle's front to determine whether the vehicle in front has started moving. Processor 101 can communicate with speaker 103, display 104, and headlights 105, and can control speaker 103, display 104, and headlights 105 to emit corresponding prompts.

[0034] Reference Figure 2 The diagram illustrates a flowchart of a method for identifying the starting of a vehicle according to an embodiment of this application.

[0035] like Figure 2 As shown, the method for recognizing the starting of the vehicle in front may include the following steps:

[0036] Step 201: After the current vehicle enters the waiting state for the vehicle in front to start, at least two frontal images of the current vehicle are acquired.

[0037] When a vehicle is in motion, if it encounters traffic jams or stops at a red light, it will change from a moving state to a stationary state, thus entering a waiting state for the vehicle in front to start. Therefore, in one optional implementation, it is possible to determine in real time or at regular intervals whether the vehicle has changed from a moving state to a stationary state, and when it is determined that the vehicle has entered a waiting state for the vehicle in front to start.

[0038] If the current vehicle is stationary when it starts, it will also enter a waiting state for the vehicle in front to start. Therefore, in one optional implementation, it is possible to determine in real time or periodically whether the current vehicle is stationary when it starts, and when it is determined that the current vehicle is stationary when it starts, it is determined that the current vehicle enters a waiting state for the vehicle in front to start.

[0039] For example, the current vehicle speed can be obtained through the speed sensor on the vehicle. When the current vehicle speed is greater than 0, the current vehicle is determined to be in a moving state; when the current vehicle speed is 0, the current vehicle is determined to be in a stationary state.

[0040] For example, the gear information of the current vehicle can be obtained. When the gear information of the current vehicle is forward (D), it can be determined that the current vehicle is in motion. When the gear information of the current vehicle is neutral (N) or park (P), it can be determined that the current vehicle is stationary.

[0041] After determining that the current vehicle has entered a waiting state for the vehicle in front to start moving, the camera can be controlled to capture at least two images of the current vehicle's front. For example, the camera can capture at least two images of the current vehicle's front at preset time intervals. The specific time interval can be set based on practical experience, and this embodiment does not impose any limitations on it.

[0042] Step 202: Obtain at least two target images based on the at least two front images.

[0043] In one alternative implementation, for each of the at least two forward images, the entire forward image can be considered as a target image to obtain the at least two target images. That is, one forward image is considered as one target image. By employing this method, no further processing of the acquired forward images is required; the brake lights on the vehicle ahead of the current vehicle can be directly determined from the at least two forward images (i.e., the target images) to whether they have changed from illuminated to off.

[0044] In one optional implementation, for each of the at least two forward images, a target region is determined in the forward image, and an image within the target region is cropped from the forward image as the target image, thereby obtaining the at least two target images. The target region includes the brake lights of the vehicle ahead. By employing this method, the acquired forward images are further processed, and the target region containing the brake lights is cropped as the target image. Based on the cropped target image, it is determined whether the brake lights of the vehicle ahead of the current vehicle have changed from on to off, thereby reducing workload and improving processing efficiency. The target region can refer to a Region of Interest (ROI).

[0045] For example, the area where the vehicle is located in the foreground image can be determined, and the area where the vehicle is located can be used as the target area.

[0046] For example, when determining the area where a vehicle is located in the image ahead, a preset vehicle detection model can be used to detect vehicles in the image ahead, obtain a vehicle detection box in the image ahead, and determine the area bounded by the vehicle detection box as the area where the vehicle is located.

[0047] For example, a vehicle detection model can be pre-trained to detect vehicles in images.

[0048] The training process for a vehicle detection model may include:

[0049] First, construct the model to be trained. For example, the model to be trained can be any neural network model with object detection function. The model to be trained can include, but is not limited to: R-CNN (Region Convolutional Neural Network) series models, SSD (Single Shot MultiBox Detector) series models, YOLO (You Only Look Once) series models, etc.

[0050] Then, sample data is acquired. For example, the sample data may include a sample image and labeling information of the sample image, wherein the labeling information of the sample image is used to indicate the actual type of each pixel in the sample image, wherein the type includes vehicle and non-vehicle.

[0051] Next, the model to be trained is trained using sample data, and the trained model is used as the vehicle detection model. Specifically, the sample images are used as input to the model to be trained, and the output of the model is the predicted type of each pixel in the sample image. Based on the output of the model to be trained and the labeling information of the sample images, the loss function of the model is calculated. If the loss function is less than a preset loss threshold, training is considered complete. For example, the loss function may include, but is not limited to, cross-entropy loss function, absolute value loss function, squared loss function, etc.

[0052] Therefore, the process of using a preset vehicle detection model to detect vehicles in the image ahead and obtaining vehicle detection boxes in the image ahead may include: inputting the image ahead into a pre-trained vehicle detection model, obtaining the output of the vehicle detection model, wherein the output of the vehicle detection model is the type of each pixel in the image ahead, and then obtaining vehicle detection boxes in the image ahead based on the type of each pixel in the image ahead.

[0053] For example, if the camera on the current vehicle was installed before the vehicle left the factory, a fixed area can be pre-defined in the captured forward image as the forward vehicle image area corresponding to the current vehicle, based on the relative positional relationship between the current vehicle's hood line position information, the vehicle's centerline position information, and the camera's position information. Therefore, when determining the location of the vehicle in the forward image, the pre-defined forward vehicle image area corresponding to the current vehicle can be obtained as the location of the vehicle in the forward image.

[0054] For example, if the camera on the current vehicle is installed after the vehicle leaves the factory, when determining the area where the vehicle is located in the forward image, the position information of the hood line, the position information of the vehicle's center line, and the position information of the camera can be obtained, and the area where the vehicle is located in the forward image can be determined according to the relative positional relationship between the hood line, the center line, and the camera.

[0055] For example, the hood line of the current vehicle can be used as the bottom edge, and the intersection of the vehicle's centerline and the hood line can be used as the midpoint of the bottom edge. Based on the relative positional relationships between the hood line position information, the vehicle centerline position information, and the camera position information, a region with a preset distance, preset length, and preset width in front of the current vehicle can be selected as the area where the vehicle is located in the foreground image. The specific values ​​of the preset distance, preset length, and preset width can be set based on practical experience (such as vehicle width, vehicle height, etc.), and this embodiment does not impose any restrictions on this. For example, the preset distance can be set to 2m, the preset length to 2m, and the preset width to 2m, etc.

[0056] For example, the region where the vehicle is located in the forward image can be determined, and the region where the brake lights are located in the forward image can be determined based on the region where the vehicle is located, and the region where the brake lights are located can be used as the target region. By determining the region where the brake lights are located, the image region to be analyzed can be further precisely located.

[0057] Reference Figure 3 The diagram illustrates a flowchart of an embodiment of this application for determining the area where the brake light is located.

[0058] like Figure 3 As shown, the process of determining the location of the brake lights in the forward image based on the area where the vehicle is located may include the following steps A1 to A3:

[0059] Step A1: Identify the vehicle model in the image ahead.

[0060] For example, a preset vehicle model recognition model can be used to identify the vehicle model in the forward image to obtain the vehicle model in the forward image. In practical applications, any applicable vehicle model recognition model can be selected to identify the vehicle model in the acquired forward image, which will not be discussed in detail in this embodiment.

[0061] For example, the license plate number in the image ahead can be identified, and then the vehicle model corresponding to the license plate number can be retrieved from a preset database. The retrieved vehicle model is then identified as the vehicle model in the image ahead. In practical applications, any applicable method can be used to identify the license plate number in the image ahead; this embodiment will not elaborate on this further. For instance, a license plate number recognition model can be used to identify the license plate number in the image ahead, thereby obtaining the license plate number in the image ahead.

[0062] Step A2: Obtain the brake light position information corresponding to the vehicle model.

[0063] After obtaining the vehicle model in the forward image, the brake light position information corresponding to that vehicle model can be retrieved from the vehicle information stored in the database. The brake light position information may include the position of the brake light relative to the rear of the vehicle, or the height of the brake light relative to the bottom of the vehicle, or the vertical distance of the brake light relative to the side of the vehicle.

[0064] Step A3: Based on the vehicle's location area and the brake light's position information, determine the area where the brake lights are located in the forward image.

[0065] After obtaining the location of the vehicle in the foreground image and the position information of the corresponding brake lights, the location of the brake lights in the foreground image can be determined from the vehicle's location area. By determining the location of the brake lights, the image area to be analyzed can be further precisely located, and the area requiring image processing can be further reduced, thereby reducing the computational load.

[0066] Step 203: Based on the at least two target images, determine whether the brake lights on the vehicle in front of the current vehicle have changed from being on to being off.

[0067] Reference Figure 4 The diagram illustrates a flowchart of an embodiment of this application for determining whether the brake light on the vehicle in front of the current vehicle has changed from being lit to being off, based on at least two target images.

[0068] like Figure 4 As shown, the process of determining whether the brake light on the vehicle in front of the current vehicle has changed from on to off, based on the at least two target images, may include the following steps B1 to B2:

[0069] Step B1: For each target image, determine candidate pixels in the target image and extract the target channel pixel value of the candidate pixels, wherein the target channel is an image channel related to the state change of the brake light.

[0070] For example, considering that the state change of the brake light involves a change in the brake light's color—bright red when the brake light is on and dark red when it is off—the red channel pixel value of the brake light will undergo a significant change (e.g., decrease) when the brake light changes from on to off. In other words, the red channel is the image channel related to the brake light's state change. Therefore, when the at least two target images are in RGB (Red, Green, Blue) format, each pixel in the target images can be identified as a candidate pixel, and the R channel pixel value of the candidate pixel can be extracted as the target channel pixel value of the candidate pixel.

[0071] For example, considering that the state change of the brake light involves the brightness change of the brake light, the brake light is brighter when it is on and lower when it is off. This results in a significant change (e.g., a decrease) in the brightness channel pixel value of the brake light when it changes from on to off. In other words, the brightness channel is the image channel related to the state change of the brake light. Therefore, when the at least two target images are in RGB format, each RGB target image can be first converted to HSV (Hue, Saturation, Value) format. Then, each pixel in the HSV format target image can be determined as a candidate pixel, and the V channel pixel value of the candidate pixel can be extracted as the target channel pixel value of the candidate pixel.

[0072] For example, considering that the change in brake light state involves a change in brake light brightness, and further taking into account that in an HSV format image, the H, S, and V channels each have their own corresponding threshold ranges when the main color of the brake light (i.e., red) is present, that is, when red is present, the pixel values ​​of the H channel have a corresponding red threshold range, the pixel values ​​of the S channel also have a corresponding red threshold range, and the pixel values ​​of the V channel also have a corresponding red threshold range. Therefore, pixels in the target image whose H channel pixel values, S channel pixel values, and V channel pixel values ​​satisfy the preset threshold range of the main color of the brake light (i.e., red) can be determined as candidate pixels, and the V channel pixel values ​​of the candidate pixels are extracted as the target channel pixel values ​​of the candidate pixels. In this way, compared to the method in the above embodiment where every pixel in the target image is determined as a candidate pixel, the number of candidate pixels can be further reduced by limiting the threshold range, thereby further reducing the amount of computation. The above threshold ranges can be set according to actual experience, and this embodiment does not impose any restrictions on them.

[0073] Step B2: Based on the changes in the target channel pixel values ​​of the candidate pixels in the at least two target images, determine whether the brake lights on the vehicle in front of the current vehicle have changed from being on to being off.

[0074] Reference Figure 5 This document illustrates a flowchart of an embodiment of the present application for determining whether the brake light on the vehicle in front of the current vehicle has changed from on to off based on the change in the target channel pixel value of candidate pixels in at least two target images.

[0075] like Figure 5 As shown, the process of determining whether the brake light on the vehicle in front of the current vehicle has changed from on to off based on the changes in the target channel pixel values ​​of candidate pixels in the at least two target images may include the following steps B21 to B24:

[0076] Step B21: Use the first frame of the target image among the at least two target images as a reference image.

[0077] Step B22: For each candidate pixel, calculate the difference between the target channel pixel value of the candidate pixel in the current target image and the target channel pixel value of the candidate pixel in the reference image.

[0078] Starting from the second frame of the at least two target images, for each candidate pixel in the current target image, calculate the difference between the target channel pixel value of the candidate pixel in the current target image and the target channel pixel value of the candidate pixel in the reference image (by subtracting the target channel pixel value of the candidate pixel in the reference image from the target channel pixel value of the candidate pixel in the current target image), so that a corresponding difference can be obtained for each candidate pixel in the current target image.

[0079] For example, if the current target image contains 200 candidate pixels and the reference image also contains 200 candidate pixels, then calculate the difference between the target channel pixel value of the first candidate pixel in the current target image and the target channel pixel value of the first candidate pixel in the reference image. Calculate the difference between the target channel pixel value of the second candidate pixel in the current target image and the target channel pixel value of the second candidate pixel in the reference image, and so on, until the difference between the target channel pixel value of the 200th candidate pixel in the current target image and the target channel pixel value of the 200th candidate pixel in the reference image is obtained.

[0080] Step B23: When the first target pixel can form the shape of a brake light, determine that the brake light on the vehicle in front of the current vehicle changes from lit to off. The first target pixel refers to a candidate pixel whose difference is negative and whose absolute value of the difference is greater than a preset deviation threshold.

[0081] If the difference value corresponding to a candidate pixel in the current target image is negative, it means that the target channel pixel value of the candidate pixel is smaller than the target channel pixel value of the candidate pixel in the reference image. For example, the R channel pixel value or V channel pixel value of the candidate pixel is smaller. At the same time, if the absolute value of the difference value corresponding to the candidate pixel is greater than the preset deviation threshold, it can be determined that it is caused by the brake light changing from being on to being off.

[0082] Therefore, after calculating the corresponding difference for each candidate pixel in the current target image, candidate pixels with negative differences and absolute values ​​greater than a preset deviation threshold can be selected as the first target pixel. It is then determined whether the first target pixel can form a brake light shape. If the first target pixel can form a brake light shape, it can be determined that the brake light on the vehicle in front of the current vehicle has changed from on to off. Selecting the first target pixel in this way eliminates interference from other pixel changes, improving the accuracy of recognition.

[0083] Step B24: When the second target pixel can form a brake light shape, the current target image is used as a reference image, and the process returns to step B22. The second target pixel refers to a candidate pixel whose difference is positive and whose difference is greater than the deviation threshold.

[0084] If the difference value corresponding to a candidate pixel in the current target image is positive, it means that the target channel pixel value of the candidate pixel is larger than the target channel pixel value of the candidate pixel in the reference image. For example, the R channel pixel value or V channel pixel value of the candidate pixel is larger. At the same time, if the difference value corresponding to the candidate pixel is greater than the preset deviation threshold, it can be determined that it is caused by the brake light changing from off to on.

[0085] Therefore, after calculating the corresponding difference for each candidate pixel in the current target image, candidate pixels with positive differences and differences greater than a preset deviation threshold can be used as second target pixels. It is then determined whether the second target pixel can form a brake light shape. If the second target pixel can form a brake light shape, it can be determined that the brake light on the vehicle in front of the current vehicle has changed from off to on. In this case, the current target image can be used as a reference image, and the process returns to step B22 to restart the identification of whether the brake light on the vehicle in front of the current vehicle has changed from on to off.

[0086] The specific value of the aforementioned deviation threshold can be set based on practical experience, specifically the pixel difference when the brake light changes from on to off. The aforementioned brake light shape can encompass various possible brake light shapes in practical applications. This embodiment does not impose limitations on the above parameters.

[0087] Step 204: When the brake lights on the vehicle in front change from on to off, determine that the vehicle in front has started moving.

[0088] When the brake lights on the vehicle ahead change from being on to being off, it can be determined that the vehicle ahead has started moving.

[0089] In one alternative implementation, after determining that the vehicle in front has started moving, a first prompt message can be issued to remind the driver of the current vehicle that the vehicle in front has started moving, so as to promptly remind the driver of the current vehicle that the vehicle in front has started moving and that the driver can prepare to depart.

[0090] For example, the first prompt information may include, but is not limited to, at least one of the following: light prompt information, sound prompt information, text prompt information, etc. For instance, a light prompt information may be issued by controlling the driver's seat ambient light inside the vehicle to flash, or a sound prompt information may be issued by controlling the vehicle's internal speakers to broadcast a voice message, or a text prompt information may be issued by controlling the display on the vehicle's internal display or head-up display, etc.

[0091] In one optional implementation, considering that when the vehicle in front starts moving, not only will its brake lights change from on to off, but the vehicle will also move, resulting in significant changes to the target image captured by the current vehicle, a method combining determining whether the changes between at least two target images meet preset conditions and determining whether the brake lights on the vehicle in front change from on to off can be used to determine whether the vehicle in front has started moving. When the changes between at least two target images meet the preset conditions and the brake lights on the vehicle in front change from on to off, it is determined that the vehicle in front has started moving.

[0092] For example, in the process of determining whether the change between the at least two target images meets the preset conditions, the pixel values ​​of the pixels in the current target image and the pixel values ​​of the pixels in the previous target image (the previous target image refers to the frame image before the current target image in terms of the order of acquisition time) can be obtained, and the number of pixels in the current target image whose pixel values ​​change more than a preset change threshold compared to the previous target image can be obtained. When the number of pixels exceeds the preset number threshold, it is determined that the change between the current target image and the previous target image meets the preset conditions.

[0093] Understandably, if the method of identifying the start of a vehicle in front is solely based on whether the changes between at least two target images meet preset conditions, then a large threshold value (e.g., 100) is needed to confirm the start of the vehicle. However, if a method combines determining whether the changes between at least two target images meet preset conditions with determining whether the brake lights on the vehicle in front have changed from on to off, the threshold value can be set to a smaller value (e.g., 50) than when using the single image change method. Therefore, the number of pixels requiring change is reduced, and the processing time is shortened. Thus, the combined approach improves both the accuracy and efficiency of the identification.

[0094] In this embodiment, based on the phenomenon that the brake lights on the vehicle in front change from illuminated to off when the vehicle starts moving, the system identifies the vehicle starting by determining whether the brake lights have changed from illuminated to off. On one hand, since the phenomenon of the brake lights changing from illuminated to off occurs earlier than the vehicle moving, the timeliness of identifying the vehicle starting is improved. On the other hand, the phenomenon of brake lights changing from illuminated to off is less affected by other interference factors in the image, thus improving the accuracy of identifying the vehicle starting.

[0095] Reference Figure 6 The diagram illustrates a flowchart of another method for identifying the starting of a vehicle in front, according to an embodiment of this application.

[0096] like Figure 6 As shown, the method for recognizing the starting of the vehicle in front may include the following steps:

[0097] Step 601: After the current vehicle enters the waiting state for the vehicle in front to start, at least two frontal images of the current vehicle are acquired.

[0098] Step 602: Obtain at least two target images based on the at least two front images.

[0099] Step 603: For each target image, divide the target image into a current lane image and adjacent lane images.

[0100] For example, lane line detection can be performed on the target image to obtain the lane lines contained in the target image. Then, based on these lane lines, the target image can be divided into a current lane image and adjacent lane images. The lane where the current vehicle is located is the current lane, and the two adjacent lanes to the left and right of the current lane are the adjacent lanes.

[0101] For example, a preset lane detection model can be used to detect lane lines in the target image to obtain the lane lines in the target image. In practical applications, any suitable lane detection model can be selected to detect lane lines in the target image, which will not be discussed in detail in this embodiment.

[0102] Step 604: Based on the current lane image and the adjacent lane image, determine whether the brake light on the vehicle in front of the current vehicle has changed from being on to being off.

[0103] For each target image, the current lane image and adjacent lane images can be obtained, thus obtaining at least two current lane images corresponding to the current lane, and at least two adjacent lane images corresponding to each adjacent lane.

[0104] For the current lane, it can be determined, based on at least two current lane images, whether the brake lights on the vehicle in front of the current vehicle (the vehicle in front in the current lane) have changed from being on to being off.

[0105] For each adjacent lane, it can be determined whether the brake lights on the vehicle in front of the current vehicle (the vehicle in front in the adjacent lane) have changed from being on to being off, based on the images of at least two adjacent lanes corresponding to that adjacent lane.

[0106] The specific judgment process is basically the same as the process described above for judging whether the brake light on the vehicle in front of the current vehicle has changed from lit to off based on the at least two target images. For details, please refer to the relevant descriptions in the above embodiments. This embodiment will not be discussed in detail here.

[0107] Step 605: When the brake light in at least one of the vehicles ahead of the current vehicle changes from on to off, determine that the vehicle ahead has started moving.

[0108] When the brake light of at least one vehicle ahead of the current vehicle changes from illuminated to off, it can be determined that the vehicle ahead has started moving. This method allows for more timely determination of the vehicle ahead's start, even when the vehicle ahead in the current lane has not started moving in time, while the vehicle ahead in an adjacent lane has already started moving, provided that the vehicle's starting conditions are met.

[0109] In one possible scenario, the lane types of the current lane and adjacent lanes can be further considered. If the vehicle type in the adjacent lane is the same as the current lane, then when the vehicle in front in the adjacent lane starts moving, the current vehicle in the current lane will also start moving. Therefore, in this case, when the brake light of the vehicle in front in the adjacent lane changes from illuminated to off, it can be determined that the vehicle in front has started moving. If the vehicle type in the adjacent lane is different from the current lane, then when the vehicle in front in the adjacent lane starts moving, the current vehicle in the current lane cannot start moving. Therefore, in this case, when the brake light of the vehicle in front in the adjacent lane changes from illuminated to off, it will not be determined that the vehicle in front has started moving. By taking into account the influence of lane type, the accuracy of recognition can be further improved.

[0110] The lane types mentioned above may include, but are not limited to: left-turn lane, straight-ahead lane, right-turn lane, left-turn-straight-ahead lane, right-turn-straight-ahead lane, etc. Among them, the left-turn lane and the left-turn-straight-ahead lane have the same lane type, the right-turn lane and the right-turn-straight-ahead lane have the same lane type, and the straight-ahead lane and the left-turn-straight-ahead lane and the right-turn-straight-ahead lane have the same lane type.

[0111] Step 606: When the brake light on the vehicle ahead in the adjacent lane image changes from lit to off, a second prompt message is issued to prompt the current vehicle to change lanes, and / or a third prompt message is issued to prompt the vehicle ahead to start moving.

[0112] In one optional implementation, when it is determined that the brake light on the vehicle ahead in the adjacent lane image changes from illuminated to off, especially if the vehicle ahead in the current lane has not yet started moving, a second prompt message can be issued to remind the driver of the current vehicle to change lanes, thereby avoiding obstruction from the vehicle ahead in the current lane. Exemplarily, the second prompt message may include, but is not limited to, at least one of the following: light prompts, sound prompts, text prompts, etc.

[0113] In one optional implementation, when it is determined that the brake light on the vehicle ahead in the adjacent lane image changes from on to off, especially if the vehicle ahead in the current lane has not yet started moving, a third prompt message can be issued to remind the vehicle ahead to start moving, so as to promptly remind the vehicle ahead in the current lane to start moving, avoid blocking the lane, and ensure the smooth progress of the current vehicle. For example, the third prompt message may include, but is not limited to, at least one of the following: light prompt message, sound prompt message, text prompt message, etc.

[0114] For example, the aforementioned light-based warning messages can be displayed by controlling the flashing of the vehicle's headlights. The aforementioned sound-based warning messages can be displayed by controlling the vehicle's external speakers or external sound systems (such as external sound transducers). For text-based warning messages, text can be transmitted to the vehicle ahead via V2X (vehicle-to-everything communication technology).

[0115] Reference Figure 7 The diagram shows a structural block diagram of a vehicle start-up recognition device according to an embodiment of this application.

[0116] like Figure 7 As shown, the vehicle start recognition device includes a processor 701 and a camera 702.

[0117] The camera 702 is used to capture at least two frontal images of the current vehicle after the current vehicle enters the waiting state for the vehicle in front to start, and transmit the at least two frontal images to the processor 701;

[0118] The processor 701 includes:

[0119] Acquisition module 7011 is used to receive the at least two front images and acquire at least two target images based on the at least two front images;

[0120] The first judgment module 7012 is used to determine, based on the at least two target images, whether the brake light on the vehicle in front of the current vehicle changes from lit to off; when the brake light on the vehicle in front changes from lit to off, the vehicle in front is determined to start moving.

[0121] Optionally, the device further includes a speaker 703 for emitting a first prompt message to indicate that the vehicle in front of the current vehicle has started moving.

[0122] Optionally, the first determination module 7012 is specifically used to divide each target image into a current lane image and an adjacent lane image; determine, based on the current lane image and the adjacent lane image respectively, whether the brake light on the vehicle in front of the current vehicle changes from lit to off; and determine that the vehicle in front starts moving when the brake light on at least one vehicle in front of the current vehicle changes from lit to off.

[0123] Optionally, the device further includes a speaker 703; the speaker 703 is configured to emit a second prompt message to prompt the current vehicle to change lanes when the brake light on the vehicle ahead in the adjacent lane image changes from lit to off, and / or to emit a third prompt message to prompt the vehicle ahead to start moving.

[0124] Optionally, the acquisition module 7011 includes: a first acquisition unit 70111, configured to, for each of the forward images, use the forward image as the target image to obtain the at least two target images; or, a second acquisition unit 70112, configured to, for each of the forward images, determine a target region in the forward image, and extract an image within the target region from the forward image as the target image to obtain the at least two target images; the target region includes the brake light.

[0125] Optionally, the first judgment module 7012 includes: a pixel extraction unit 70121, configured to determine candidate pixels in each target image and extract the target channel pixel value of the candidate pixels, wherein the target channel is an image channel related to the state change of the brake light; and a pixel judgment unit 70122, configured to determine whether the brake light on the vehicle in front of the current vehicle has changed from lit to off based on the change of the target channel pixel value of the candidate pixels in the at least two target images.

[0126] Optionally, the pixel extraction unit 70121 is specifically used to determine each pixel in the target image as the candidate pixel, and extract the red R channel pixel value of the candidate pixel as the target channel pixel value of the candidate pixel; or, determine each pixel in the target image as the candidate pixel, and extract the brightness V channel pixel value of the candidate pixel as the target channel pixel value of the candidate pixel; or, determine the pixels in the target image whose hue H channel pixel value, saturation S channel pixel value, and brightness V channel pixel value satisfy a preset brake light main color threshold range as the candidate pixel, and extract the V channel pixel value of the candidate pixel as the target channel pixel value of the candidate pixel.

[0127] Optionally, the pixel determination unit 70122 is specifically used to take the first frame of the target image in the at least two target images as a reference image; for each candidate pixel, calculate the difference between the target channel pixel value of the candidate pixel in the current target image and the target channel pixel value of the candidate pixel in the reference image; when the first target pixel can form a brake light shape, determine that the brake light on the vehicle in front of the current vehicle changes from on to off, wherein the first target pixel refers to a candidate pixel whose difference is negative and whose absolute value of the difference is greater than a preset deviation threshold; when the second target pixel can form a brake light shape, take the current target image as a reference image and return to execute the step of calculating the difference between the target channel pixel value of the candidate pixel in the current target image and the target channel pixel value of the candidate pixel in the reference image for each candidate pixel, wherein the second target pixel refers to a candidate pixel whose difference is positive and whose difference is greater than the deviation threshold.

[0128] Optionally, the second acquisition unit 70112 is specifically used to determine the area where the vehicle is located in the forward image, and to use the area where the vehicle is located as the target area.

[0129] The second acquisition unit 70112 is further specifically used to determine the area where the vehicle is located in the forward image; based on the area where the vehicle is located, determine the area where the brake lights are located in the forward image, and take the area where the brake lights are located as the target area.

[0130] Optionally, the second acquisition unit 70112 is specifically used to perform vehicle detection on the front image using a preset vehicle detection model, obtain a vehicle detection box in the front image, and determine the vehicle detection box as the area where the vehicle is located.

[0131] Optionally, the second acquisition unit 70112 is specifically used to acquire a preset image region of the vehicle in front corresponding to the current vehicle, as the region where the vehicle is located in the image in front.

[0132] Optionally, the second acquisition unit 70112 is specifically used to acquire the hood line position information, vehicle centerline position information, and camera position information of the current vehicle; and to determine the area where the vehicle is located in the forward image based on the relative positional relationship between the hood line position information, the vehicle centerline position information, and the camera position information.

[0133] Optionally, the second acquisition unit 70112 is specifically used to identify the vehicle model in the forward image; acquire the brake light position information corresponding to the vehicle model; and determine the area where the brake light is located in the forward image based on the area where the vehicle is located and the brake light position information.

[0134] Optionally, the processor 701 further includes: a second judgment module 7013, used to judge whether the current vehicle has changed from a driving state to a stationary state; when the current vehicle changes from a driving state to a stationary state, it is determined that the current vehicle has entered a waiting state for the preceding vehicle to start.

[0135] Optionally, the processor 701 further includes a third judgment module 7014, used to judge whether the changes between the at least two target images meet preset conditions; when the preset conditions are met, it is determined that the vehicle in front has started moving.

[0136] In this embodiment, based on the phenomenon that the brake lights on the vehicle in front change from illuminated to off when the vehicle starts moving, the system identifies the vehicle starting by determining whether the brake lights have changed from illuminated to off. On one hand, since the phenomenon of the brake lights changing from illuminated to off occurs earlier than the vehicle moving, the timeliness of identifying the vehicle starting is improved. On the other hand, the phenomenon of brake lights changing from illuminated to off is less affected by other interference factors in the image, thus improving the accuracy of identifying the vehicle starting.

[0137] In embodiments of this application, an electronic device is also provided. This electronic device may include one or more processors and one or more computer-readable storage media storing instructions thereon, such as an application program. When the instructions are executed by the one or more processors, the processors cause the processors to perform the preceding vehicle start-up recognition method of any of the above embodiments.

[0138] Figure 8 A schematic diagram of the structure of an electronic device 800 according to an embodiment of this application is shown. Figure 8 As shown, the electronic device 800 includes a Central Processing Unit (CPU) 801, which can perform various appropriate actions and processes based on computer program instructions stored in Read Only Memory (ROM) 802 or loaded from storage unit 808 into Random Access Memory (RAM) 803. The RAM 803 can also store various programs and data required for the operation of the electronic device 800. The CPU 801, ROM 802, and RAM 803 are interconnected via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0139] Multiple components in electronic device 800 are connected to I / O interface 805, including: input unit 806, such as keyboard, mouse, microphone, etc.; output unit 807, such as various types of displays, speakers, etc.; storage unit 808, such as disk, optical disk, etc.; and communication unit 809, such as network card, modem, wireless transceiver, etc. Communication unit 809 allows electronic device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0140] The various processes and handling described above can be executed by the processing unit 801. For example, the preceding vehicle start-up recognition method of any of the above embodiments can be implemented as a computer software program, which is tangibly contained in a computer-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by CPU 801, one or more actions in the preceding vehicle start-up recognition method described above can be performed.

[0141] In embodiments of this application, a computer-readable storage medium is also provided, on which a computer program is stored, which can be executed by a processor of an electronic device, and when the computer program is executed by the processor, the processor performs the preceding vehicle start-up recognition method as described in any of the above embodiments.

[0142] The processors mentioned above may include, but are not limited to: CPU, Network Processor (NP), Digital Signal Processor (DSP), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0143] The computer-readable storage media mentioned above may include, but are not limited to: ROM, RAM, Compact Disc Read Only Memory (CD-ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), Hard Disk, Floppy Disk, Flash Memory, etc.

[0144] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0145] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0146] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0147] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

[0148] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0149] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0150] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0151] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0152] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0153] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0154] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. In summary, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for recognizing the starting of a vehicle ahead, characterized in that, The method includes: After the current vehicle enters the waiting state for the vehicle in front to start, at least two frontal images of the current vehicle are acquired; At least two target images are obtained based on the at least two front images; Based on the at least two target images, determine whether the brake lights on the vehicle in front of the current vehicle have changed from being on to being off; When the brake lights on the vehicle ahead change from lit to off, it is determined that the vehicle ahead has started moving. The step of determining whether the brake lights on the vehicle in front of the current vehicle have changed from on to off based on the at least two target images includes: For each target image, candidate pixels in the target image are determined, and the target channel pixel value of the candidate pixels is extracted, wherein the target channel is an image channel related to the state change of the brake light; The first frame of the target image among the at least two target images is used as the reference image; For each candidate pixel, calculate the difference between the target channel pixel value of the candidate pixel in the current target image and the target channel pixel value of the candidate pixel in the reference image; When the first target pixel can form the shape of a brake light, it is determined that the brake light on the vehicle in front of the current vehicle changes from being lit to being off. The first target pixel refers to a candidate pixel whose difference is negative and whose absolute value of the difference is greater than a preset deviation threshold.

2. The method according to claim 1, wherein acquiring at least two target images based on the at least two forward images comprises: For each of the foreground images, the foreground image is used as the target image to obtain the at least two target images; or, For each of the forward images, a target region in the forward image is determined, and an image within the target region is cropped from the forward image as the target image, resulting in at least two target images; the target region includes the brake light.

3. The method according to claim 1, wherein determining candidate pixels in the target image and extracting target channel pixel values ​​of the candidate pixels comprises: Each pixel in the target image is determined as a candidate pixel, and the red R channel pixel value of the candidate pixel is extracted as the target channel pixel value of the candidate pixel. or, Each pixel in the target image is determined as a candidate pixel, and the brightness V channel pixel value of the candidate pixel is extracted as the target channel pixel value of the candidate pixel. or, The pixels in the target image whose hue (H channel), saturation (S channel), and luminance (V channel) values ​​satisfy a preset threshold range for the main color of the brake light are identified as candidate pixels. The V channel pixel value of the candidate pixel is then extracted as the target channel pixel value of the candidate pixel.

4. The method according to claim 1, further comprising: When the second target pixel can form a brake light shape, the current target image is used as a reference image, and the step of calculating the difference between the target channel pixel value of the candidate pixel in the current target image and the target channel pixel value of the candidate pixel in the reference image is returned for each candidate pixel. The second target pixel refers to a candidate pixel whose difference is positive and whose difference is greater than the deviation threshold.

5. The method according to claim 2, wherein determining the target region in the foreground image comprises: The region where the vehicle is located in the foreground image is determined, and the region where the vehicle is located is taken as the target region.

6. The method according to claim 2, wherein determining the target region in the forward image comprises: Determine the area where the vehicle is located in the image ahead; Based on the area where the vehicle is located, the area where the brake lights are located in the forward image is determined, and the area where the brake lights are located is taken as the target area.

7. The method according to claim 5 or 6, wherein determining the area where the vehicle is located in the forward image includes: The vehicle detection model is used to detect vehicles in the image ahead, and vehicle detection boxes are obtained in the image ahead. The vehicle detection boxes are then identified as the regions where the vehicles are located.

8. The method according to claim 5 or 6, wherein determining the area where the vehicle is located in the forward image comprises: Obtain the preset image region of the vehicle ahead corresponding to the current vehicle, and use it as the region where the vehicle is located in the image ahead.

9. The method according to claim 5 or 6, wherein determining the area where the vehicle is located in the forward image comprises: Obtain the current vehicle's hood line position information, vehicle centerline position information, and camera position information; Based on the relative positional relationship between the hood line position information, the vehicle centerline position information, and the camera position information, the area where the vehicle is located in the forward image is determined.

10. The method according to claim 6, wherein determining the location of the brake lights in the forward image based on the location of the vehicle comprises: Identify the vehicle model in the forward image; Obtain the brake light location information corresponding to the vehicle model; Based on the vehicle's location area and the brake light's position information, the area where the brake lights are located in the forward image is determined.

11. The method according to claim 1, further comprising, after determining that the vehicle in front has started moving: Issue a first prompt message to indicate that the vehicle in front of the current vehicle has started moving.

12. The method according to claim 1, The step of determining whether the brake lights on the vehicle in front of the current vehicle have changed from on to off based on the at least two target images includes: For each target image, the target image is divided into a current lane image and adjacent lane images; Based on the current lane image and the adjacent lane image respectively, determine whether the brake light on the vehicle in front of the current vehicle has changed from being on to being off; Determining that the vehicle in front starts moving when the brake light on the vehicle in front changes from lit to off includes: determining that the vehicle in front starts moving when the brake light on at least one of the vehicles in front of the current vehicle changes from lit to off.

13. The method of claim 12, further comprising, after determining that the vehicle ahead has started moving: When the brake light on the vehicle ahead in the adjacent lane image changes from lit to off, a second prompt message is issued to prompt the current vehicle to change lanes, and / or a third prompt message is issued to prompt the vehicle ahead to start moving.

14. The method according to claim 1, further comprising, before acquiring the forward image of the current vehicle after the current vehicle enters the waiting state for the preceding vehicle to start, the method further comprising: Determine whether the current vehicle has transitioned from a moving state to a stationary state; When the current vehicle changes from a moving state to a stationary state, it is determined that the current vehicle enters a waiting state for the vehicle in front to start.

15. The method according to claim 1, further comprising: Determine whether the changes between the at least two target images meet preset conditions; When the preset conditions are met, the vehicle in front is determined to start moving.

16. A device for recognizing when a vehicle starts moving ahead, characterized in that, The device includes a processor and a camera; The camera is used to capture at least two frontal images of the current vehicle after the current vehicle enters a waiting state for the vehicle in front to start, and to transmit the at least two frontal images to the processor; The processor includes: The acquisition module is configured to receive the at least two front images and acquire at least two target images based on the at least two front images; The first judgment module is used to determine, based on the at least two target images, whether the brake light on the vehicle in front of the current vehicle changes from lit to off; when the brake light on the vehicle in front changes from lit to off, the vehicle in front is determined to start moving. The first judgment module includes: A pixel extraction unit is used to determine candidate pixels in each target image and extract the target channel pixel value of the candidate pixels, wherein the target channel is an image channel related to the state change of the brake light. A pixel determination unit is used to take the first frame of the target image in the at least two target images as a reference image; for each candidate pixel, calculate the difference between the target channel pixel value of the candidate pixel in the current target image and the target channel pixel value of the candidate pixel in the reference image; when the first target pixel can form a brake light shape, determine that the brake light on the vehicle in front of the current vehicle changes from on to off, wherein the first target pixel refers to a candidate pixel whose difference is negative and whose absolute value of the difference is greater than a preset deviation threshold.

17. The apparatus of claim 16, further comprising a speaker for emitting a first prompt message to indicate that the vehicle in front of the current vehicle has started moving.

18. The apparatus according to claim 16, wherein the first determining module is specifically configured to, for each target image, divide the target image into a current lane image and an adjacent lane image; determine, based on the current lane image and the adjacent lane image respectively, whether the brake light on the vehicle in front of the current vehicle changes from lit to off; and determine that the vehicle in front starts moving when the brake light on at least one vehicle in front of the current vehicle changes from lit to off.

19. The apparatus of claim 18, further comprising a loudspeaker; The speaker is configured to emit a second prompt message to indicate that the current vehicle is changing lanes when the brake light on the vehicle ahead in the adjacent lane image changes from lit to off, and / or to emit a third prompt message to indicate that the vehicle ahead is starting to move.

20. An electronic device, characterized in that, include: One or more processors; and One or more computer-readable storage media on which instructions are stored; When the instruction is executed by the one or more processors, the processors perform the preceding vehicle start-up recognition method as described in any one of claims 1 to 15.

21. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, causes the processor to perform the preceding vehicle start-up recognition method as described in any one of claims 1 to 15.

Images