A vehicle image data processing method and device

Abstract

The application discloses a vehicle image data processing method and device. The method comprises the following steps: after obtaining multiple images corresponding to a target vehicle at a target time point, performing splicing processing on the multiple images to obtain a target splicing image corresponding to the target vehicle at the target time point, performing image compression processing on the target splicing image, and transmitting the compressed data. The method fuses multiple image compression processes of multiple cameras corresponding to the target vehicle into a single image compression process, simplifies the image compression process in the image acquisition process of the intelligent vehicle, and can effectively reduce the hardware resource and computing power consumption of the image compression process. Furthermore, the method fuses multiple image encoding stream transmission processes of multiple cameras corresponding to the vehicle into a single image encoding stream transmission process, which can effectively reduce the complexity of the image transmission process of the intelligent vehicle and reduce the network resource consumption of the image uploading process of the intelligent vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent vehicles, in particular to a vehicle image data processing method and device, an electronic device and a computer readable storage medium. The present application also relates to a vehicle image data response method. BACKGROUND

[0002] In the process of automatic driving of an intelligent vehicle, image data needs to be collected to meet the needs of formulating or optimizing an automatic driving strategy. In this process, the images captured by multiple cameras of the vehicle need to be compressed and encoded respectively, and the compressed code streams need to be transmitted, for example, uploaded to a cloud server, so that the cloud server can train a vehicle automatic driving strategy model based on the decoded image data to obtain a suitable automatic driving strategy for the vehicle or optimize and update an existing automatic driving strategy for the vehicle. However, compressing and encoding the images captured by each camera of the vehicle respectively and uploading them respectively makes the image compression and transmission process more complex and requires higher hardware resources and computing power. SUMMARY

[0003] The embodiments of the present application provide a vehicle image data processing method, device, electronic device and computer readable storage medium to solve the problem that in the prior art, the images captured by each camera of the vehicle are compressed and encoded respectively and uploaded respectively, which makes the image compression and transmission process more complex and requires higher hardware resources and computing power.

[0004] The embodiments of the present application provide a vehicle image data processing method applied to a vehicle-mounted image processing device of an intelligent vehicle, which comprises the following steps:

[0005] Obtaining multiple images corresponding to a target vehicle at a target time point;

[0006] Performing stitching processing on the multiple images to obtain a target stitched image corresponding to the target vehicle at the target time point;

[0007] Performing image compression processing on the target stitched image and transmitting the compressed data.

[0008] In an embodiment, the step of performing stitching processing on the multiple images to obtain a target stitched image corresponding to the target vehicle at the target time point comprises merging the multiple images into the target stitched image.

[0009] In an embodiment, the step of merging the multiple images into the target stitched image comprises:

[0010] Obtaining preset orientation information between multiple cameras of the target vehicle;

[0011] According to the preset orientation information, the images taken by each camera are combined and arranged to obtain the target spliced image.

[0012] In an embodiment, the merging of the multiple images into the target spliced image comprises:

[0013] Images containing the same taken content in the multiple images are taken as adjacent images, and the adjacent images are combined and arranged to obtain the target spliced image.

[0014] In an embodiment, the image compression processing of the target spliced image comprises:

[0015] The same taken content contained in the multiple images is taken as spatial redundancy information of the target spliced image, and the spatial redundancy information is used for intra-frame encoding.

[0016] In an embodiment, the transmission of the compressed data comprises:

[0017] The compressed data is sent to a cloud server, so that the cloud server decodes the compressed data to obtain the target spliced image, and performs model training on an automatic driving strategy model corresponding to the target vehicle based on the target spliced image.

[0018] According to another aspect of the present application, a vehicle image data response method is provided, which is applied to a cloud server, and the method comprises:

[0019] Receiving an image encoding stream from a target vehicle;

[0020] Decoding the image encoding stream, and obtaining a target spliced image corresponding to the target vehicle at a target time point from the decoded image, the target spliced image being an image obtained by splicing multiple images corresponding to the target vehicle at the target time point;

[0021] Analyzing an automatic driving strategy corresponding to the target vehicle based on the target spliced image.

[0022] According to another aspect of the present application, a vehicle image data processing device is provided, which is applied to a vehicle-mounted image processing equipment of an intelligent vehicle, and the device comprises

[0023] An image acquisition unit is configured to acquire multiple images corresponding to a target vehicle at a target time point;

[0024] An image splicing unit is configured to splice the multiple images to obtain a target spliced image corresponding to the target vehicle at the target time point;

[0025] An image compression and transmission unit performs image compression processing on the target spliced image and transmits the compressed data.

[0026] According to another aspect of the present application, an electronic device is provided, comprising a processor and a memory; wherein the memory is configured to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the above method.

[0027] According to another aspect of the present application, a computer readable storage medium is provided, which stores one or more computer instructions, wherein the instructions are executed by a processor to implement the above method.

[0028] Compared with the prior art, the present application has the following advantages:

[0029] The vehicle image data processing method provided by the present application comprises the following steps: obtaining a plurality of images corresponding to a target vehicle at a target time point; splicing the plurality of images to obtain a target spliced image corresponding to the target vehicle at the target time point; performing image compression processing on the target spliced image; and transmitting the compressed data. The method fuses the image compression processes of the plurality of images corresponding to the plurality of cameras of the target vehicle into a single image compression process, simplifies the image compression process in the image acquisition process of the intelligent vehicle, and can effectively reduce the hardware resource and computing power consumption of the image compression process. Furthermore, the present application fuses the image encoding and streaming transmission processes of the plurality of images corresponding to the plurality of cameras of the vehicle into a single image encoding and streaming transmission process, which can effectively reduce the complexity of the image transmission process of the intelligent vehicle and reduce the network resource consumption of the image uploading process of the intelligent vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 is a flow chart of the vehicle image data processing method provided by an embodiment of the present application;

[0031] Figure 2 is a flow chart of the vehicle image data response method provided by an embodiment of the present application;

[0032] Figure 3 is a unit block diagram of the vehicle image data processing device provided by an embodiment of the present application;

[0033] Figure 4 is a unit block diagram of the vehicle image data response device provided by an embodiment of the present application;

[0034] Figure 5 is a logical structure schematic diagram of the electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0035] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, the present application can be practiced without the specific details. In other instances, well-known methods have not been described in detail in order not to unnecessarily obscure aspects of the present application.

[0036] In the automatic driving scene of the intelligent vehicle, image data acquisition needs to be performed to meet the data analysis needs of the automatic driving strategy, such as meeting the formulation or optimization needs of the automatic driving strategy. In this process, the images captured by multiple cameras of the vehicle need to be compressed and encoded respectively, and the compressed code streams need to be transmitted, for example, uploaded to a cloud server, so that the cloud server can train the vehicle automatic driving strategy model based on the decoded image data to obtain the applicable automatic driving strategy of the vehicle or optimize and update the existing automatic driving strategy of the vehicle. However, the compression and encoding of the images captured by each camera of the vehicle respectively and the uploading respectively make the image compression and transmission process more complex and require higher hardware resources and computing power.

[0037] In view of the intelligent vehicle image acquisition scene, in order to avoid the problem that the images captured by each camera of the vehicle are compressed and encoded respectively and uploaded respectively in the prior art, which makes the image compression and transmission process more complex and requires higher hardware resources, computing power, and network resources, the present application provides a vehicle image data processing method, a vehicle image data processing device corresponding to the method, an electronic device, and a computer readable storage medium. The present application also provides a vehicle image data response method, a device corresponding to the method, an electronic device, and a computer readable storage medium. The embodiments are provided below to describe the above method, device, electronic device, and computer readable storage medium in detail.

[0038] An embodiment of the present application provides a vehicle image data processing method. The application subject of the method can be a vehicle-mounted image processing application, which can run on a vehicle-mounted image processing device of an intelligent vehicle. Figure 1 The flowchart of the vehicle image data processing method provided by the present embodiment is as follows, which is described in detail below Figure 1 The method provided by the present embodiment is described in detail. The embodiments described below are used to explain the principles of the method and are not limited in actual use.

[0039] As shown in Figure 1 The vehicle image data processing method provided by the present embodiment includes the following steps:

[0040] S101, obtaining multiple images corresponding to a target vehicle at a target time point.

[0041] The step is used for acquiring multiple images corresponding to the target vehicle at a target time point. For example, in the automatic driving process of the intelligent automobile, multiple images taken by the vehicle-mounted image processing device of multiple cameras of the target vehicle at the target time point are collected. The target vehicle can be provided with cameras in each direction, which are used to take images from each angle, which can represent the driving conditions and the surrounding environment conditions of the target vehicle at each time point. Each camera can be distributed on the target vehicle based on preset orientation information. In this embodiment, the multiple images corresponding to the target time point can be extracted from the video data taken by each camera. Each camera can also take images based on a preset time interval to obtain the multiple images corresponding to the preset target time point. The target time point can refer to a time point, or a plurality of preset time points. For example, multiple images taken by the multiple cameras at each of the plurality of time points are acquired respectively.

[0042] It should be noted that the multiple images corresponding to the target time point of the target vehicle can be taken by multiple cameras of the target vehicle, or can be taken by other cameras, such as cameras of associated vehicles of the target vehicle and / or cameras of the highway. The associated vehicles of the target vehicle can be vehicles traveling in the same direction or in the opposite direction of the target vehicle. For example, at the target time point, the target vehicle meets another vehicle, and at this time, a camera is provided on the roadside. The camera of the vehicle meeting the target vehicle and / or the camera of the roadside facility take images of the driving conditions and the surrounding environment of the target vehicle, and transmit the taken multiple images to the vehicle-mounted image processing device of the target vehicle based on short-time networking or other short-distance communication between vehicles and roadside facilities.

[0043] S102, performing stitching processing on the multiple images to obtain a target stitching image corresponding to the target vehicle at the target time point.

[0044] After obtaining the plurality of images corresponding to the target vehicle at the target time point in the above step, the plurality of images are stitched to obtain a target stitched image corresponding to the target vehicle at the target time point. The stitching of the plurality of images can be implemented in various ways. For example, the plurality of images can be selectively stitched based on whether the same object is included in each image. For example, image 1, image 2, and image 3 in the plurality of images all include one or more same objects (e.g., the same pedestrian crossing), and image 1, image 2, and image 3 are stitched. For another example, the plurality of images can be stitched according to the similarity between the images, and images with high similarity are stitched. For another example, the plurality of images can be stitched based on the proportion of the same background region included in each image. For example, image 1, image 2, and image 3 all include background region A, and the proportion of background region A in image 1, image 2, and image 3 all exceeds 50%, and image 1, image 2, and image 3 are stitched. The above stitching of the plurality of images based on whether the same object is included in each image, based on the similarity between the images, and based on the proportion of the same background region included in each image all aim to increase the proportion of the same factors in the target stitched image after stitching, thereby facilitating subsequent image compression processing of the target stitched image.

[0045] It should be noted that the above stitching of the plurality of images can be performed after cutting each image. For example, the same background part included in each image can be cut off, and only the different background part of each image is retained, and each image corresponding to a different background is stitched. Alternatively, the importance of the content of each image can be divided, the unimportant image content can be cut off, only the important image content is retained, and the important image content is stitched. In this embodiment, in order to simplify the image stitching process and improve the image stitching efficiency, the plurality of images can not be subjected to additional processing such as cutting, and only simple merging is performed, i.e., the plurality of images are merged into a target stitched image, for example, the plurality of images are arranged and combined into one image in any manner.

[0046] It should be noted that when the above multiple images are taken by multiple cameras of the target vehicle, considering that the positions of the cameras can be preset, for example, the cameras are arranged on the body of the target vehicle according to a predetermined field of view angle or a predetermined interval, the multiple images can be combined into the target spliced image in the following manner: first, preset position information between the multiple cameras of the target vehicle is obtained, for example, a first camera is located at a first position and is used to take an image in a 0-degree to 40-degree field of view range with the body center as the origin, a second camera is located at a second position and is used to take an image in a 40-degree to 80-degree range with the body center as the origin, and a third camera is located at a third position and is used to take an image in an 80-degree to 120-degree range with the body center as the origin; then, the images taken by the cameras are arranged according to the preset position information to obtain the target spliced image, for example, the images taken by the first camera, the second camera, and the third camera are arranged in sequence from left to right.

[0047] In another embodiment, when the positions of the cameras are not preset, the images can also be arranged based on whether the images contain the same taken content, and specifically, the images containing the same taken content in the multiple images can be arranged as adjacent images, and the adjacent images are arranged to obtain the target spliced image.

[0048] S103, performing image compression processing on the target spliced image, and transmitting the compressed data.

[0049] After the multiple images are spliced in the above step to obtain the target spliced image corresponding to the target vehicle at the target time point, the target spliced image is subjected to image compression processing in this step, and the data obtained after compression is transmitted, so as to complete the image acquisition process of the target vehicle at the target time point. The specific manner of image compression processing of the target image can adopt an existing image encoding method, for example, an intra-frame prediction encoding method based on H264, H265, MPEG, and other video encoding standards, so as to reduce the spatial redundancy of the target spliced image. Specifically, the same taken content contained in the multiple images can be taken as spatial redundancy information of the target spliced image, and the spatial redundancy information is subjected to intra-frame encoding.

[0050] In the embodiment, the transmitting the compressed data can specifically refer to sending the compressed data to a cloud server, so that the cloud server decodes the compressed data to obtain the target spliced image, and analyzes the automatic driving strategy of the target vehicle based on the target spliced image, for example, taking the target spliced image as a training sample to train the automatic driving strategy model corresponding to the target vehicle, so as to obtain the automatic driving strategy suitable for the target vehicle or optimize and update the existing automatic driving strategy of the target vehicle.

[0051] The vehicle image data processing method provided by the embodiment can be applied to a vehicle-mounted image processing device of an intelligent vehicle. After a plurality of images corresponding to a target vehicle at a target time point are obtained, the plurality of images are spliced to obtain a target spliced image corresponding to the target vehicle at the target time point, the target spliced image is subjected to image compression processing, and the compressed data is transmitted. The method fuses the multiple image compression processes of the plurality of cameras corresponding to the target vehicle into a single image compression process by splicing the plurality of images corresponding to the target vehicle at the same time point, simplifies the image compression process in the image acquisition process of the intelligent vehicle, and can effectively reduce the hardware resource consumption and computing power consumption of the image compression process. Moreover, the application fuses the multiple image encoding stream transmission processes of the plurality of cameras of the vehicle into a single image encoding stream transmission process, which can effectively reduce the complexity of the image transmission process of the intelligent vehicle and reduce the network resource consumption of the image uploading process of the intelligent vehicle.

[0052] The above embodiment provides a vehicle image data processing method. Correspondingly, another embodiment of the application further provides a vehicle image data response method. The vehicle image data response method can be applied to a cloud server. The following describes Figure 2 The vehicle image data response method provided by the embodiment is described in detail. The embodiments described below are used to explain the principles of the method and are not limited in actual use. As shown in the following description, Figure 2 The vehicle image data response method includes the following steps.

[0053] S201, receiving an image encoding stream from a target vehicle.

[0054] This step is used to receive an image encoding stream from a target vehicle, for example, receiving an image encoding stream processed by the steps S101, S102 and S103 of the above embodiment.

[0055] S202, decoding the image encoding stream and obtaining a target spliced image corresponding to the target vehicle at a target time point from the decoded image.

[0056] The target spliced image is an image obtained by splicing a plurality of images corresponding to the target vehicle at the target time point, for example, after the vehicle image processing device obtains a plurality of images corresponding to the target vehicle at the target time point (captured by a plurality of cameras of the target vehicle, or captured by a camera preset by an associated vehicle of the target vehicle and / or a highway), the plurality of images are spliced to obtain a target spliced image corresponding to the target vehicle at the target time point, for example, selective image splicing can be performed based on whether the same object is included in each image, for example, image 1, image 2, and image 3 in the plurality of images all include one or more same objects (for example, including the same pedestrian crossing), then image 1, image 2, and image 3 are spliced; for another example, the images can be spliced according to the similarity between the images, and the images with high similarity are spliced; for another example, the images can be spliced based on the proportion of the same background area included in each image, for example, image 1, image 2, and image 3 all include background area A, and the proportion of A background area in image 1, image 2, and image 3 all exceeds 50%, then image 1, image 2, and image 3 are spliced. The purpose of splicing based on whether the same object is included in each image, based on the similarity between the images, and based on the proportion of the same background area included in each image is to increase the proportion of the same factors in the spliced target spliced image, so as to improve the encoding efficiency when the target spliced image is compressed. In addition, the process of splicing the plurality of images can be to splice after cutting each image, for example, cutting the same background part included in each image, only keeping the different background parts of each image, and splicing each image corresponding to different backgrounds; or dividing the importance of the content in each image, cutting the unimportant image content, only keeping the important image content, and splicing the important image content. In order to simplify the image splicing process and improve the image splicing efficiency, the plurality of images can not be cut and subjected to additional processing, but only be simply combined, that is, the plurality of images are combined into a target spliced image, for example, the plurality of images are arranged and combined into one image in any manner.It should be noted that when the above multiple images are taken by multiple cameras of the target vehicle, considering that the positions of the cameras can be pre-set, for example, the cameras are arranged around the target vehicle according to a predetermined field of view angle or a predetermined distance, the above multiple images can be combined into a target splicing image in the following manner: first, obtain preset position information between the multiple cameras of the target vehicle, for example, a first camera is located at a first position and is used to take images within a 0-degree to 40-degree field of view with the center of the vehicle body as the origin, a second camera is located at a second position and is used to take images within a 40-degree to 80-degree range with the center of the vehicle body as the origin, and a third camera is located at a third position and is used to take images within an 80-degree to 120-degree range with the center of the vehicle body as the origin; then, according to the above preset position information, the images taken by the above cameras are arranged in combination to obtain a target splicing image, for example, the images taken by the above first camera, second camera and third camera are arranged in combination from left to right. In the case where the positions of the cameras are not pre-set, the images can also be arranged in combination based on whether they contain the same captured content. Specifically, the images containing the same captured content in the above multiple images can be regarded as adjacent images, and the adjacent images are arranged in combination to obtain a target splicing image.

[0057] S203, analyzing an automatic driving strategy corresponding to the target vehicle based on the target splicing image.

[0058] After the above step of decoding the image encoding stream and obtaining the target splicing image corresponding to the target vehicle at the target time point from the decoded image, this step is used to analyze the automatic driving strategy corresponding to the target vehicle based on the target splicing image, for example, taking the target splicing image as a training sample, model training is performed on the automatic driving strategy model corresponding to the target vehicle to obtain an automatic driving strategy suitable for the target vehicle or to optimize and update the existing automatic driving strategy of the target vehicle.

[0059] The vehicle image data response method provided in this embodiment first receives an image encoding stream from a target vehicle; decodes the image encoding stream and obtains a target splicing image corresponding to the target vehicle at a target time point from the decoded image, the target splicing image being an image obtained by splicing multiple images corresponding to the target vehicle at the target time point; finally, analyzing an automatic driving strategy corresponding to the target vehicle based on the target splicing image. By using this method, after the target vehicle uploads its target splicing image to the cloud server, the cloud server does not need to decompress the image encoding stream corresponding to each camera separately, and does not need to perform timestamp alignment on each image encoding stream when analyzing the data of the automatic driving strategy corresponding to the target vehicle, which can effectively reduce the computing power consumption of the cloud server and improve the data analysis efficiency.

[0060] The above embodiment provides a vehicle image data processing method. Correspondingly, another embodiment of the present application further provides a vehicle image data processing device. The device is applied to a vehicle-mounted image processing equipment of an intelligent vehicle. Since the device embodiment is basically similar to the method embodiment, the description is relatively simple. For details of the related technical features, please refer to the corresponding description of the above-provided method embodiment. The following description of the device embodiment is only illustrative.

[0061] For reference Figure 3 To understand this embodiment, Figure 3 The unit block diagram of the vehicle image data processing device provided in this embodiment is shown in Figure 3 The vehicle image data processing device provided in this embodiment includes:

[0062] An image acquisition unit 301 is configured to acquire a plurality of images corresponding to a target vehicle at a target time point.

[0063] An image stitching unit 302 is configured to perform stitching processing on the plurality of images to obtain a target stitched image corresponding to the target vehicle at the target time point.

[0064] An image compression and transmission unit 303 is configured to perform image compression processing on the target stitched image and transmit the compressed data.

[0065] In an implementation, the stitching processing on the plurality of images to obtain the target stitched image corresponding to the target vehicle at the target time point includes merging the plurality of images into the target stitched image.

[0066] In an implementation, the merging of the plurality of images into the target stitched image includes:

[0067] Obtaining preset orientation information between a plurality of cameras of the target vehicle.

[0068] According to the preset orientation information, combining and arranging each image taken by each camera to obtain the target stitched image.

[0069] In an implementation, the merging of the plurality of images into the target stitched image includes:

[0070] Taking images containing the same taken content in the plurality of images as adjacent images, and combining and arranging the adjacent images to obtain the target stitched image.

[0071] In an implementation, the image compression processing on the target stitched image includes:

[0072] The same shooting content contained in the plurality of images is taken as spatial redundancy information of the target spliced image, and intra-frame encoding is performed based on the spatial redundancy information.

[0073] In an implementation, the transmitting the compressed data comprises:

[0074] The compressed data is sent to a cloud server, so that the cloud server decodes the compressed data to obtain the target spliced image, and performs model training on an automatic driving strategy model corresponding to the target vehicle based on the target spliced image.

[0075] The vehicle image data processing device provided by the embodiment of the present application fuses the multiple image compression processes of the multiple cameras corresponding to the target vehicle into a single image compression process by splicing the multiple images corresponding to the target vehicle at the same time point and performing image compression processing on the spliced target spliced image. This simplifies the image compression process in the image acquisition process of the intelligent vehicle and can effectively reduce the hardware resource and computing power consumption of the image compression process. In addition, the device fuses the multiple image encoding stream transmission processes of the multiple cameras of the vehicle into a single image encoding stream transmission process, which can effectively reduce the complexity of the image transmission process of the intelligent vehicle and reduce the network resource consumption of the image uploading process of the intelligent vehicle.

[0076] In the above embodiment, a vehicle image data processing method and a vehicle image data processing device are provided. In addition, another embodiment of the present application also provides an electronic device. Since the electronic device embodiment is basically similar to the method embodiment, it is described more simply. For details of related technical features, please refer to the corresponding description of the above-provided method embodiment. The following description of the electronic device embodiment is only illustrative. The electronic device embodiment comprises the following:

[0077] For details, please refer to Figure 5 For understanding the embodiment, Figure 5 The electronic device provided by the embodiment is shown in the schematic diagram.

[0078] As Figure 5 shown, the electronic device provided by the embodiment comprises a processor 501 and a memory 502.

[0079] The memory 502 is used to store computer instructions for data processing. When the computer instructions are read and executed by the processor 501, the following operations are performed:

[0080] Obtain multiple images corresponding to a target vehicle at a target time point;

[0081] Splice the multiple images to obtain a target spliced image corresponding to the target vehicle at the target time point.

[0082] performing image compression processing on the target spliced image, and transmitting compressed data.

[0083] In an embodiment, the splicing processing of the multiple images to obtain the target spliced image corresponding to the target vehicle at the target time point comprises: merging the multiple images into the target spliced image.

[0084] In an embodiment, the merging of the multiple images into the target spliced image comprises:

[0085] obtaining preset orientation information between multiple cameras of the target vehicle;

[0086] According to the preset orientation information, each image taken by each camera is combined and arranged to obtain the target spliced image.

[0087] In an embodiment, the merging of the multiple images into the target spliced image comprises:

[0088] The images containing the same taken content in the multiple images are regarded as adjacent images, and the adjacent images are combined and arranged to obtain the target spliced image.

[0089] In an embodiment, the image compression processing of the target spliced image comprises:

[0090] The same taken content contained in the multiple images is regarded as spatial redundancy information of the target spliced image, and the spatial redundancy information is intra-coded.

[0091] In an embodiment, the transmitting of the compressed data comprises:

[0092] The compressed data is sent to a cloud server, so that the cloud server decodes the compressed data to obtain the target spliced image, and trains a model of an automatic driving strategy of the target vehicle based on the target spliced image.

[0093] The electronic device provided in the embodiment fuses the multiple image compression processes of the multiple cameras corresponding to the target vehicle into a single image compression process by performing image stitching on the multiple images corresponding to the target vehicle at the same time point and performing image compression processing on the target stitched image after stitching, simplifies the image compression process in the image acquisition process of the intelligent vehicle, and can effectively reduce the hardware resource and computing power consumption of the image compression process. In addition, the electronic device fuses the multiple image encoding and streaming processes of the multiple cameras of the vehicle into a single image encoding and streaming process, which can effectively reduce the complexity of the image transmission process of the intelligent vehicle and reduce the network resource consumption of the image uploading process of the intelligent vehicle.

[0094] In the above embodiment, a vehicle image data processing method, a vehicle image data processing device, and an electronic device are provided. In addition, another embodiment of the present application also provides a computer readable storage medium for implementing the above method. The computer readable storage medium embodiment provided in the present application is described relatively simply, and the related parts can be referred to the corresponding description of the above method embodiment. The following described embodiments are only illustrative.

[0095] The computer readable storage medium provided in the embodiment stores computer instructions, and the instructions are executed by a processor to implement the following steps:

[0096] Obtain multiple images corresponding to a target vehicle at a target time point;

[0097] Perform stitching processing on the multiple images to obtain a target stitched image corresponding to the target vehicle at the target time point;

[0098] Perform image compression processing on the target stitched image and transmit the compressed data.

[0099] In an implementation manner, the performing stitching processing on the multiple images to obtain the target stitched image corresponding to the target vehicle at the target time point includes merging the multiple images into the target stitched image.

[0100] In an implementation manner, the merging the multiple images into the target stitched image includes:

[0101] Obtain preset orientation information between multiple cameras of the target vehicle;

[0102] According to the preset orientation information, combine and arrange each image taken by each camera to obtain the target stitched image.

[0103] In an implementation manner, the merging the multiple images into the target stitched image includes:

[0104] The images containing the same shooting content in the plurality of images are adjacent images, and the adjacent images are arranged and combined to obtain the target spliced image.

[0105] In an implementation, the image compression processing of the target spliced image comprises:

[0106] The same shooting content contained in the plurality of images is spatial redundancy information of the target spliced image, and the spatial redundancy information is used for intra-frame encoding.

[0107] In an implementation, the transmission of the compressed data comprises:

[0108] The compressed data is sent to a cloud server, so that the cloud server decodes the compressed data to obtain the target spliced image, and performs model training on an automatic driving strategy model corresponding to the target vehicle based on the target spliced image.

[0109] By executing the computer instructions stored on the computer readable storage medium provided in the embodiment, the plurality of images corresponding to the target vehicle at the same time point are spliced, and the spliced target spliced image is subjected to image compression processing. This way fuses the multiple image compression processes of the plurality of cameras corresponding to the target vehicle into a single image compression process, simplifies the image compression process in the image acquisition process of the intelligent vehicle, and effectively reduces the hardware resource and computing power consumption of the image compression process. And the multiple image encoding stream transmission processes corresponding to the plurality of cameras of the vehicle are fused into a single image encoding stream transmission process, which can effectively reduce the complexity of the image transmission process of the intelligent vehicle and reduce the network resource consumption of the image uploading process of the intelligent vehicle.

[0110] The above embodiment provides a vehicle image data response method. Correspondingly, another embodiment of the present application also provides a vehicle image data response device. The device is applied to a cloud server. Since the device embodiment is basically similar to the method embodiment, the description is relatively simple. For details of related technical features, please refer to the corresponding description of the above-provided method embodiment. The following description of the device embodiment is only illustrative.

[0111] For reference Figure 4 To understand this embodiment, Figure 4 The unit block diagram of the vehicle image data response device provided in the embodiment is shown in Figure 4 The vehicle image data response device provided in the embodiment comprises:

[0112] The image encoding stream receiving unit 401 is configured to receive the image encoding stream from the target vehicle.

[0113] The decoding unit 402 is configured to decode the image encoding stream, and obtain a target spliced image corresponding to the target vehicle at a target time point from the decoded image, the target spliced image being an image obtained by splicing a plurality of images corresponding to the target vehicle at the target time point.

[0114] The automatic driving strategy analysis unit 403 is configured to analyze an automatic driving strategy corresponding to the target vehicle based on the target spliced image.

[0115] By using the vehicle image data response apparatus provided in the embodiment, after the target vehicle uploads the target spliced image to the cloud server, the cloud server does not need to decompress the image encoding stream corresponding to each camera respectively, and does not need to perform timestamp alignment on the image encoding stream when performing data analysis on the automatic driving strategy corresponding to the target vehicle, so that the computing power consumption of the cloud server can be effectively reduced, and the data analysis efficiency is improved.

[0116] In one typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0117] The memory can include non-persistent memory in the computer readable medium, random access memory (RAM), and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer readable media.

[0118] 1. Computer readable media includes permanent and non-permanent, removable and non-removable media, which can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device, or any other non-transmission medium that can be used to store information accessible by a computing device. According to the definition in this paper, computer readable media does not include non-transitory computer readable media (transitory media), such as modulated data signals and carriers.

[0119] 2. Those skilled in the art will appreciate that embodiments of the present application can be readily used as a method, a system or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer-readable program code.

[0120] Although the present application has been disclosed in connection with the preferred embodiments thereof, it should be understood that many modifications, variations and additions therein can be made by those skilled in the art and it is intended to cover all such modifications, variations and additions in the scope of the present application. Accordingly, the protection scope of the present application should be limited by the scope of the claims.

Claims

1. A vehicle image data processing method, characterized by, A vehicle-mounted image processing device applied to an intelligent vehicle, the method comprising: obtaining a plurality of images corresponding to a target vehicle at a target time point; performing stitching processing on the plurality of images to obtain a target stitched image corresponding to the target vehicle at the target time point, comprising: selecting at least two images containing the same content in the plurality of images as adjacent images, wherein the at least two images also each contain the same background part; based on the adjacent images, arranging the adjacent images in combination, cutting the same background part contained in each of the adjacent images, and merging the cut adjacent images into the target stitched image; performing image compression processing on the target stitched image, and transmitting the compressed data, comprising: sending the compressed data to a cloud server for decoding the compressed data by the cloud server to obtain the target stitched image, and analyzing an automatic driving strategy corresponding to the target vehicle based on the target stitched image.

2. The method of claim 1, wherein, The image compression processing on the target stitched image comprises: regarding the same content contained in the plurality of images as spatial redundancy information of the target stitched image, and performing intra-frame encoding based on the spatial redundancy information.

3. A vehicle image data response method, characterized by, A method applied to a cloud server, the method comprising: receiving an image encoding stream from a target vehicle; decoding the image encoding stream, and obtaining a target stitched image corresponding to the target vehicle at a target time point from the decoded image, the target stitched image being an image obtained by performing stitching processing on a plurality of images corresponding to the target vehicle at the target time point, the stitching comprising: selecting at least two images containing the same content in the plurality of images as adjacent images, wherein the at least two images also each contain the same background part, arranging the adjacent images in combination based on the adjacent images, cutting the same background part contained in each of the adjacent images, and merging the cut adjacent images into the target stitched image; analyzing an automatic driving strategy corresponding to the target vehicle based on the target stitched image.

4. A vehicle image data processing apparatus characterized by comprising: A vehicle-mounted image processing device applied to an intelligent vehicle, the device comprising an image acquisition unit configured to obtain a plurality of images corresponding to a target vehicle at a target time point; an image stitching unit configured to perform stitching processing on the plurality of images to obtain a target stitched image corresponding to the target vehicle at the target time point, comprising: selecting at least two images containing the same content in the plurality of images as adjacent images, wherein the at least two images also each contain the same background part, arranging the adjacent images in combination based on the adjacent images, cutting the same background part contained in each of the adjacent images, and merging the cut adjacent images into the target stitched image; an image compression and transmission unit configured to perform image compression processing on the target stitched image, and transmit the compressed data, comprising: sending the compressed data to a cloud server for decoding the compressed data by the cloud server to obtain the target stitched image, and analyzing an automatic driving strategy corresponding to the target vehicle based on the target stitched image.

5. An electronic device, comprising: comprising a processor and a memory; wherein the memory is to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method of any one of claims 1-3.

6. A computer-readable storage medium having stored thereon one or more computer instructions, wherein, the instructions are executed by the processor to implement the method of any one of claims 1-3.

Images