A DVR image processing method, device and medium for reducing CPU load

Abstract

The present invention discloses a DVR image processing method, device and medium for reducing CPU load, the method comprising: acquiring a camera image of a target vehicle; parsing the camera image, copying the parsed camera image through a preset interface of a graphics processor in the target vehicle to obtain a target image; encoding preset key information in the target image to generate a static image; merging the static image with a dynamic image in the target image, and playing the merged image. The present invention proposes a DVR image processing method, device and medium for reducing CPU load, using a graphics processor to replace the CPU for image processing tasks, which can reduce the burden of the CPU, and only cropping and encoding key information, so that the amount of data to be processed can be significantly reduced, thereby improving processing efficiency, and can solve the problem of difficulty in reducing CPU load when processing and playing DVR images of vehicles.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a DVR image processing method, device and medium for reducing CPU load. Background Art

[0002] In the field of vehicle driving, on-board cameras play a pivotal role. They are an indispensable part of the core sensing equipment. By capturing and analyzing image information, on-board cameras can accurately identify vehicles, pedestrians, traffic signs and various obstacles on the road, providing extremely rich and detailed environmental information for vehicle driving. Therefore, these sophisticated devices can provide drivers with a full range of 360° visual perception, so that the details of the vehicle's surrounding environment are under control. The existing third-party app image processing methods for automotive vehicle systems mainly use overlay array coding to copy and crop images, so that the images are stored for playback.

[0003] However, overlay array encoding requires complex encoding processing of the original image data, which involves a large amount of computing and storage operations, thereby consuming more CPU resources; and due to the complexity of the encoding process, this method often takes longer time for image processing and playback, resulting in a slower response speed of the vehicle system. Therefore, traditional DVR image processing methods consume more CPU resources. Summary of the invention

[0004] The present invention provides a DVR image processing method, device and medium for reducing CPU load, so as to solve the problem that it is difficult to reduce CPU load when processing and playing DVR images of a vehicle.

[0005] Obtain camera images of the target vehicle;

[0006] Parsing the camera image, and copying the parsed camera image through a preset interface of a graphics processor in the target vehicle to obtain a target image;

[0007] Encoding preset key information in the target image to generate a static image;

[0008] The static image and the dynamic image in the target image are combined, and the combined image is played.

[0009] The present invention copies the parsed camera image through the preset interface of the graphics processor, which fully utilizes the advantages of the graphics processor in image processing; compared with the CPU, the graphics processor is more efficient in processing a large amount of data in parallel, and can significantly reduce the CPU load; because the graphics processor undertakes most of the image processing tasks, the CPU only needs to process a small amount of control logic and data transmission tasks, thereby reducing the burden on the CPU and improving the performance of the overall system. Encode the preset key information in the target image to generate a static image. This step extracts the key information in the image and presents it in the form of a static image, reducing redundant information in the dynamic image. Compared with the prior art, the present invention uses a graphics processor to replace the CPU to perform image processing tasks, which can reduce the burden on the CPU. Only the key information is cropped and encoded, which can significantly reduce the amount of data to be processed, thereby improving processing efficiency. Therefore, it can solve the problem of difficulty in reducing the CPU load when processing and playing the vehicle's DVR image.

[0010] As a preferred solution, the camera image is analyzed as follows:

[0011] Expanding the camera image and capturing the expanded image into a binocular view and a surround view; wherein the camera image includes original images from a binocular fisheye camera and an ultra-wide-angle fisheye camera;

[0012] The binocular view and the surround view are stitched together, and the stitched image is transmitted to the vehicle computer bottom layer of the target vehicle for parsing through a preset protocol to obtain a parsed camera image.

[0013] This preferred solution can fully utilize the original image information captured by the camera by expanding, processing and intercepting the image, and generate views from different perspectives to meet the needs of different driving scenarios; among them, the binocular view can provide stereoscopic visual information of the road ahead, which helps the driver to better judge the distance and obstacles and improve driving safety; the surround view can provide a panoramic view around the vehicle, helping the driver to fully understand the surrounding environment during driving and avoid collisions. By splicing the binocular view and the surround view, the processing and storage requirements for multiple independent images can be reduced, thereby optimizing system performance and reducing hardware costs.

[0014] As a preferred solution, the target image is obtained by copying the parsed camera image through the preset interface of the graphics processor in the target vehicle, specifically:

[0015] Copying the parsed camera image through a preset interface of a graphics processor in the target vehicle to obtain a first image;

[0016] Splitting the first image into a series of two-dimensional YUV arrays to obtain an image array set;

[0017] The image array set is arranged according to preset horizontal and vertical coordinates to obtain the target image.

[0018] In this preferred solution, the traditional method is to copy through the CPU. This solution copies the image by calling the preset interface of the graphics processor, which can fully utilize the advantage of the graphics processor that can handle multiple tasks simultaneously and significantly improve the speed of image processing. Therefore, transferring the image copy task to the graphics processor can reduce the burden on the CPU and enable it to focus more on other more important tasks.

[0019] As a preferred solution, the preset key information in the target image is encoded to generate a static image, specifically:

[0020] Cut out preset key information from the target image to obtain a key information label set; wherein the preset key information includes date, time and gear position;

[0021] The key information tag set is encoded to generate the static image.

[0022] In this preferred solution, the traditional array coverage coding processing method is to cover or replace certain values ​​in the array to achieve a specific coding goal; this solution only cuts and encodes key information, so it can significantly reduce the amount of data to be processed, thereby improving processing efficiency. In addition, the key information label set in the encoded static image can provide a strong data basis for subsequent operations.

[0023] As a preferred solution, the static image and the dynamic image in the target image are merged, and the merged image is played, specifically:

[0024] Merging the static image with the dynamic image in the target image to obtain a merged image;

[0025] Marking the merged image with independent key frames to obtain a final image;

[0026] If the target vehicle is in a forward gear, playing the binocular view in the final image;

[0027] If the target vehicle is in reverse gear, the surround view in the final image is played.

[0028] In this preferred solution, static images can clearly display key information, while dynamic images can capture and display the real-time operating status of the vehicle; the merged image not only contains all key information, but also maintains the real-time and dynamic nature of the dynamic image, making the information display more comprehensive and intuitive. By marking the merged image with independent key frames, precise control and editing of the image can be achieved. In addition, the automatic switching of the view mode according to the gear position of the target vehicle can provide the driver with more intimate and practical driving assistance information.

[0029] The present application also provides a DVR image processing device for reducing CPU load, comprising an image module, a copy module, an encoding module and a merging module;

[0030] Wherein, the image module is used to obtain the camera image of the target vehicle;

[0031] The copy module is used to analyze the camera image and copy the analyzed camera image through a preset interface of the graphics processor in the target vehicle to obtain a target image;

[0032] The encoding module is used to encode the preset key information in the target image to generate a static image;

[0033] The merging module is used to merge the static image with the dynamic image in the target image, and play the merged image.

[0034] As a preferred solution, the copy module includes a capture unit and a splicing unit;

[0035] The interception unit is used to expand the camera image and intercept the expanded image into a binocular view and a surround view; wherein the camera image includes original images from a binocular fisheye camera and an ultra-wide-angle fisheye camera;

[0036] The stitching unit is used to stitch the binocular view and the surround view, and transmit the stitched image to the vehicle computer bottom layer of the target vehicle for analysis through a preset protocol to obtain a analyzed camera image.

[0037] As a preferred solution, the copy module includes a copy unit, a split unit and an arrangement unit;

[0038] The copying unit is used to copy the parsed camera image through a preset interface of a graphics processor in the target vehicle to obtain a first image;

[0039] The splitting unit is used to split the first image into a series of two-dimensional YUV arrays to obtain an image array set;

[0040] The arrangement unit is used to arrange the image array set according to preset horizontal and vertical coordinates to obtain the target image.

[0041] As a preferred solution, the encoding module includes a cropping unit and an encoding unit;

[0042] The cropping unit is used to crop preset key information from the target image to obtain a key information tag set; wherein the preset key information includes date, time and gear position;

[0043] The encoding unit is used to encode the key information tag set to generate the static image.

[0044] As a preferred solution, the merging module includes a merging unit, a key frame unit, a forward unit and a reverse unit;

[0045] Wherein, the merging unit is used to merge the static image and the dynamic image in the target image to obtain a merged image;

[0046] The key frame unit is used to add independent key frames to the combined image to obtain a final image;

[0047] The forward unit is used to play the binocular view in the final image if the target vehicle is in a forward gear;

[0048] The reversing unit is used to play the surround view in the final image if the target vehicle is in a reverse gear.

[0049] The present application also provides a storage medium, on which a computer program is stored. The computer program is called and executed by a computer to implement the DVR image processing method for reducing CPU load as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flowchart of a DVR image processing method for reducing CPU load provided in an embodiment of the present application;

[0051] Figure 2 It is a data processing logic diagram provided by an embodiment of the present application;

[0052] Figure 3 It is a structural schematic diagram of a DVR image processing device for reducing CPU load provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0054] In the description of this application, it should be understood that the term "first" is used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "several" means two or more.

[0055] A DVR image processing method for reducing CPU load provided in an embodiment of the present application is mainly used in situations where it is necessary to reduce CPU load and process and play vehicle DVR images without sacrificing DVR image quality and playback fluency.

[0056] Embodiment 1:

[0057] See also Figure 1 The embodiment of the present application provides a DVR image processing method for reducing CPU load, including S1 to S4, and the specific implementation steps are as follows:

[0058] S1. Obtain the camera image of the target vehicle.

[0059] Step S1 of the embodiment of the present application is specifically as follows:

[0060] The camera image of the target vehicle is obtained by installing a binocular fisheye camera and an ultra-wide-angle fisheye camera on the target vehicle; among them, the ultra-wide-angle fisheye camera is a front, back, left, and right surround fisheye camera.

[0061] S2, parsing the camera image, and copying the parsed camera image through a preset interface of a graphics processor in the target vehicle to obtain a target image.

[0062] Step S2 of the embodiment of the present application includes S2.1 to S2.2, specifically:

[0063] S2.1. The camera image is expanded by the domain controller of the target vehicle to eliminate distortion and restore the original shape and proportion of the image, and the expanded image is intercepted into a binocular view and a surround view; wherein the camera image includes the original image from the binocular fisheye camera and the ultra-wide-angle fisheye camera, and the resolution of the binocular view and the surround view is 1920*1080; and the "domain controller" is a key component in the modern automotive electronic architecture, which is responsible for integrating and managing multiple electronic control units (ECUs) to improve the intelligence and automation level of the vehicle;

[0064] The binocular view and the surround view are stitched into a 3840*1080 UYVY422 format image, and the stitched image is transmitted to the target vehicle's vehicle computer bottom layer for analysis at a rate of 20 frames per second through the GMSL1 protocol to obtain the analyzed camera image; the full name of "GMSL1 protocol" is "Gigabit Multimedia Serial Link 1", which is a high-speed, low-latency serial communication protocol.

[0065] In this embodiment S2.1, by expanding, processing and intercepting the images, the original image information captured by the camera can be fully utilized to generate views of different perspectives, thereby meeting the needs of different driving scenarios; among them, the binocular view can provide stereoscopic visual information of the road ahead, which helps the driver to better judge the distance and obstacles and improve driving safety; the surround view can provide a panoramic view around the vehicle, helping the driver to fully understand the surrounding environment during driving and avoid collisions. By splicing the binocular view and the surround view, the processing and storage requirements for multiple independent images can be reduced, thereby optimizing system performance and reducing hardware costs.

[0066] S2.2, passing the parsed camera image to the application layer software of the target vehicle, and controlling the DVRapp application to receive the image; wherein the "DVRapp application" is an application specially designed for a driving recorder or a vehicle video recording system;

[0067] Copying the parsed camera image through a C2D interface of a graphics processing unit (GPU) in the target vehicle to obtain an underlying image, thereby obtaining a first image;

[0068] Split the first image into a long string of two-dimensional YUV arrays to obtain an image array set;

[0069] Extract the target image array currently needed from the image array set; control the DVRapp application to create one or more application threads to specifically handle the image display task, and in the newly created application threads, arrange the elements of the target image array in order according to the horizontal and vertical coordinate information of the target image array to obtain the target image.

[0070] It should be noted that in the application embodiment, "DVR image" refers to the video image received by the DVR app application and prepared for storage or playback after a series of processing, cropping and encoding.

[0071] In this embodiment S2.2, the traditional method is to copy through the CPU. This solution calls the preset interface of the graphics processor to copy the image, which can fully utilize the advantage of the graphics processor that can handle multiple tasks simultaneously and significantly improve the speed of image processing; therefore, transferring the image copy task to the graphics processor can reduce the burden on the CPU and enable it to focus more on other more important tasks.

[0072] S3. Encode the preset key information in the target image to generate a static image.

[0073] Step S3 of the embodiment of the present application is specifically as follows:

[0074] Crop preset key information from the target image to obtain a key information label set; wherein the preset key information includes date, time, gear, speed, throttle and brake, etc.;

[0075] The key information tag set is encoded, a static image is generated, and the static image is stored in a specified memory.

[0076] In this embodiment S3, the traditional array coverage coding processing method is to cover or replace certain values ​​in the array to achieve a specific coding goal; this solution only cuts and encodes key information, so it can significantly reduce the amount of data to be processed, thereby improving processing efficiency. In addition, the key information label set in the encoded static image can provide a strong data basis for subsequent operations.

[0077] S4, merging the static image with the dynamic image in the target image, and playing the merged image.

[0078] Step S4 of the embodiment of the present application is specifically as follows:

[0079] Merge the static image and the dynamic image in the target image at an image recording frequency of 20 frames per second to obtain a merged image;

[0080] Use H.264 encoding to save the merged image in MP4 format, and add an independent key frame (I frame) to the merged image to ensure that the key frame can be decoded and displayed normally during the subsequent playback process to obtain the final image;

[0081] If the target vehicle is in the forward gear (D gear), the DVR app is controlled to crop the binocular view in the final image, and the cropped binocular view is stored and played;

[0082] If the target vehicle is in reverse gear (R gear), the DVR app is controlled to crop the surround view in the final image, and the cropped surround view is stored and played.

[0083] To apply the embodiments of this application, please refer to Figure 2 , Figure 2 It is a data processing logic diagram provided in an embodiment of the present application, which shows the general process of processing the camera image of the target vehicle in an embodiment of the present application.

[0084] In this embodiment S4, the static image can clearly display the key information, while the dynamic image can capture and display the real-time running status of the vehicle; the merged image not only contains all the key information, but also maintains the real-time and dynamic nature of the dynamic image, making the information display more comprehensive and intuitive;

[0085] Moreover, by marking the merged image with independent keyframes, precise control and editing of the image can be achieved, ensuring that the video can be decoded and displayed normally during subsequent playback, which can effectively reduce the CPU computing power required for playback decoding. In addition, the automatic switching of the view mode according to the gear position of the target vehicle can provide the driver with more intimate and practical driving assistance information.

[0086] Overall, this embodiment has the following beneficial effects:

[0087] This application uses the preset interface of the graphics processor to copy the parsed camera image, which fully utilizes the advantages of the graphics processor in image processing; compared with the CPU, the graphics processor is more efficient when processing large amounts of data in parallel, which can significantly reduce the CPU load; because the graphics processor undertakes most of the image processing tasks, the CPU only needs to handle a small amount of control logic and data transmission tasks, thereby reducing the burden on the CPU and improving the performance of the overall system. Encode the preset key information in the target image to generate a static image. This step extracts the key information in the image and presents it in the form of a static image, reducing redundant information in the dynamic image. Compared with the prior art, this application uses a graphics processor to replace the CPU for image processing tasks, which can reduce the burden on the CPU. Only cropping and encoding key information can significantly reduce the amount of data to be processed, thereby improving processing efficiency. Therefore, it can solve the problem of difficulty in reducing the CPU load when processing and playing the vehicle's DVR images.

[0088] In summary, this application cleverly utilizes the powerful resources of the GPU and innovatively reshapes the image merging function, thereby significantly reducing the burden on the CPU; this measure not only optimizes the video processing link of the DVR, but also effectively reduces the image quality loss and significantly improves the smoothness of video playback.

[0089] Embodiment 2:

[0090] See also Figure 3 , an embodiment of the present application provides a DVR image processing device for reducing CPU load, including an image module 10, a copy module 20, an encoding module 30 and a merging module 40;

[0091] Among them, the imaging module 10 is used to obtain the camera image of the target vehicle;

[0092] The copy module 20 is used to analyze the camera image and copy the analyzed camera image through a preset interface of the graphics processor in the target vehicle to obtain the target image;

[0093] The encoding module 30 is used to encode the preset key information in the target image to generate a static image;

[0094] The merging module 40 is used to merge the static image and the dynamic image in the target image, and play the merged image.

[0095] In one embodiment, the imaging module 10 is specifically:

[0096] The camera image of the target vehicle is obtained by installing a binocular fisheye camera and an ultra-wide-angle fisheye camera on the target vehicle; among them, the ultra-wide-angle fisheye camera is a front, back, left, and right surround fisheye camera.

[0097] In one embodiment, the copy module 20 includes a capture unit, a splicing unit, a copy unit, a splitting unit, and an arranging unit;

[0098] The interception unit is used to expand the camera image through the domain controller of the target vehicle to eliminate distortion and restore the original shape and proportion of the image, and intercept the expanded image into a binocular view and a surround view; the camera image includes the original image from the binocular fisheye camera and the ultra-wide-angle fisheye camera, and the resolution of the binocular view and the surround view is 1920*1080; and the "domain controller" is a key component in the modern automotive electronic architecture, which is responsible for integrating and managing multiple electronic control units (ECUs) to improve the intelligence and automation level of the vehicle;

[0099] The stitching unit is used to stitch the binocular view and the surround view into a 3840*1080 image in UYVY422 format, and transmit the stitched image to the target vehicle's vehicle computer bottom layer for analysis at a rate of 20 frames per second through the GMSL1 protocol to obtain the analyzed camera image; the full name of "GMSL1 protocol" is "Gigabit Multimedia Serial Link 1", which is a high-speed, low-latency serial communication protocol.

[0100] The interception unit and the splicing unit of this embodiment can fully utilize the original image information captured by the camera by expanding, processing and intercepting the image, and generate views of different perspectives to meet the needs of different driving scenarios; among them, the binocular view can provide stereoscopic visual information of the road ahead, which helps the driver to better judge the distance and obstacles and improve driving safety; the surround view can provide a panoramic view around the vehicle, helping the driver to fully understand the surrounding environment during driving and avoid collisions. By splicing the binocular view and the surround view, the processing and storage requirements for multiple independent images can be reduced, thereby optimizing system performance and reducing hardware costs.

[0101] A copy unit, used to transfer the parsed camera image to the application layer software of the target vehicle and control the DVRapp application to receive the image; wherein the "DVRapp application" is an application specially designed for a driving recorder or a vehicle video recording system;

[0102] The copy unit is further used to copy the parsed camera image through a C2D interface of a graphics processing unit (GPU) in the target vehicle to obtain an underlying image, thereby obtaining a first image;

[0103] A splitting unit, used for splitting the first image into a long string of two-dimensional YUV arrays to obtain an image array set;

[0104] The arrangement unit is used to extract the target image array currently required from the image array set; control the DVRapp application to create one or more application threads to specifically handle the image display task, and in the newly created application thread, arrange the elements of the target image array in order according to the horizontal and vertical coordinate information of the target image array to obtain the target image.

[0105] It should be noted that in the application embodiment, "DVR image" refers to the video image received by the DVR app application and prepared for storage or playback after a series of processing, cropping and encoding.

[0106] In the copy unit, splitting unit and arranging unit of this embodiment, the traditional method is to copy through the CPU. This solution calls the preset interface of the graphics processor to copy the image, which can fully utilize the advantage of the graphics processor that can handle multiple tasks simultaneously and significantly improve the speed of image processing; therefore, transferring the image copy task to the graphics processor can reduce the burden on the CPU and enable it to focus more on other more important tasks.

[0107] In one embodiment, the encoding module 30 includes a cropping unit and an encoding unit;

[0108] The cropping unit is used to crop preset key information from the target image to obtain a key information label set; wherein the preset key information includes date, time, gear, speed, throttle and brake, etc.;

[0109] The encoding unit is used to encode the key information tag set, generate a static image, and store the static image in a specified memory.

[0110] In the encoding module 30 of this embodiment, the traditional array coverage encoding processing method is to cover or replace certain values ​​in the array to achieve a specific encoding goal; this solution only cuts and encodes key information, so the amount of data to be processed can be significantly reduced, thereby improving processing efficiency. In addition, the key information label set in the encoded static image can provide a strong data basis for subsequent operations.

[0111] In one embodiment, the merging module 40 includes a merging unit, a key frame unit, a forward unit, and a reverse unit;

[0112] The merging unit is used to merge the static image and the dynamic image in the target image at an image recording frequency of 20 frames per second to obtain a merged image;

[0113] The key frame unit is used to save the merged image in MP4 format using H.264 encoding, and mark the merged image with an independent key frame (I frame) to ensure that the key frame can be decoded and displayed normally in the subsequent playback process to obtain the final image;

[0114] A forward unit, used for controlling the DVR app to crop the binocular view in the final image if the target vehicle is in the forward gear (D gear), and storing and playing the cropped binocular view;

[0115] The reversing unit is used to control the DVR app to crop the surround view in the final image if the target vehicle is in the reverse gear (R gear), and to store and play the cropped surround view.

[0116] To apply the embodiments of this application, please refer to Figure 2 , Figure 2 It is a data processing logic diagram provided in an embodiment of the present application, which shows the general process of processing the camera image of the target vehicle in an embodiment of the present application.

[0117] In the merging module 40 of this embodiment, the static image can clearly display the key information, while the dynamic image can capture and display the real-time running status of the vehicle; the merged image not only contains all the key information, but also maintains the real-time and dynamic nature of the dynamic image, making the information display more comprehensive and intuitive;

[0118] Moreover, by marking the merged image with independent keyframes, precise control and editing of the image can be achieved, ensuring that the video can be decoded and displayed normally during subsequent playback, which can effectively reduce the CPU computing power required for playback decoding. In addition, the automatic switching of the view mode according to the gear position of the target vehicle can provide the driver with more intimate and practical driving assistance information.

[0119] Overall, this embodiment has the following beneficial effects:

[0120] This application uses the preset interface of the graphics processor to copy the parsed camera image, which fully utilizes the advantages of the graphics processor in image processing; compared with the CPU, the graphics processor is more efficient when processing large amounts of data in parallel, which can significantly reduce the CPU load; because the graphics processor undertakes most of the image processing tasks, the CPU only needs to handle a small amount of control logic and data transmission tasks, thereby reducing the burden on the CPU and improving the performance of the overall system. Encode the preset key information in the target image to generate a static image. This step extracts the key information in the image and presents it in the form of a static image, reducing redundant information in the dynamic image. Compared with the prior art, this application uses a graphics processor to replace the CPU for image processing tasks, which can reduce the burden on the CPU. Only cropping and encoding key information can significantly reduce the amount of data to be processed, thereby improving processing efficiency. Therefore, it can solve the problem of difficulty in reducing the CPU load when processing and playing the vehicle's DVR images.

[0121] In summary, this application cleverly utilizes the powerful resources of the GPU and innovatively reshapes the image merging function, thereby significantly reducing the burden on the CPU; this measure not only optimizes the video processing link of the DVR, but also effectively reduces the image quality loss and significantly improves the smoothness of video playback.

[0122] Embodiment three:

[0123] The embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the DVR image processing method for reducing CPU load;

[0124] Wherein, the DVR image processing method for reducing CPU load, if implemented in the form of a software functional unit and used as an independent product, can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0125] The above are preferred embodiments of the present invention. It should be noted that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A DVR image processing method for reducing CPU load, characterized in that: include: Obtain camera images of the target vehicle; Parsing the camera image, and copying the parsed camera image through a preset interface of a graphics processor in the target vehicle to obtain a target image; Encoding preset key information in the target image to generate a static image; The static image and the dynamic image in the target image are combined, and the combined image is played.

2. A DVR image processing method for reducing CPU load as claimed in claim 1, characterized in that: Parse the camera image, specifically: Expanding the camera image and capturing the expanded image into a binocular view and a surround view; wherein the camera image includes original images from a binocular fisheye camera and an ultra-wide-angle fisheye camera; The binocular view and the surround view are stitched together, and the stitched image is transmitted to the vehicle computer bottom layer of the target vehicle for parsing through a preset protocol to obtain a parsed camera image.

3. A DVR image processing method for reducing CPU load as claimed in claim 1, characterized in that: The target image is obtained by copying the parsed camera image through the preset interface of the graphics processor in the target vehicle, specifically: Copying the parsed camera image through a preset interface of a graphics processor in the target vehicle to obtain a first image; Splitting the first image into a series of two-dimensional YUV arrays to obtain an image array set; The image array set is arranged according to preset horizontal and vertical coordinates to obtain the target image.

4. A DVR image processing method for reducing CPU load as claimed in claim 1, characterized in that: Encode the preset key information in the target image to generate a static image, specifically: Cut out preset key information from the target image to obtain a key information label set; wherein the preset key information includes date, time and gear position; The key information tag set is encoded to generate the static image.

5. A DVR image processing method for reducing CPU load as claimed in claim 1, characterized in that: The static image and the dynamic image in the target image are merged, and the merged image is played, specifically: Merging the static image with the dynamic image in the target image to obtain a merged image; Marking the merged image with independent key frames to obtain a final image; If the target vehicle is in a forward gear, playing the binocular view in the final image; If the target vehicle is in reverse gear, the surround view in the final image is played.

6. A DVR image processing device for reducing CPU load, characterized in that: It includes image module, copy module, encoding module and merging module; Wherein, the image module is used to obtain the camera image of the target vehicle; The copy module is used to parse the camera image and copy the parsed camera image through a preset interface of the graphics processor in the target vehicle to obtain a target image; The encoding module is used to encode the preset key information in the target image to generate a static image; The merging module is used to merge the static image with the dynamic image in the target image, and play the merged image.

7. A DVR image processing device for reducing CPU load as claimed in claim 6, characterized in that: The copy module includes an interception unit and a splicing unit; The interception unit is used to expand the camera image and intercept the expanded image into a binocular view and a surround view; wherein the camera image includes original images from a binocular fisheye camera and an ultra-wide-angle fisheye camera; The stitching unit is used to stitch the binocular view and the surround view, and transmit the stitched image to the vehicle computer bottom layer of the target vehicle for analysis through a preset protocol to obtain a analyzed camera image.

8. A DVR image processing device for reducing CPU load as claimed in claim 6, characterized in that: The copy module includes a copy unit, a split unit and an arrangement unit; The copying unit is used to copy the parsed camera image through a preset interface of a graphics processor in the target vehicle to obtain a first image; The splitting unit is used to split the first image into a series of two-dimensional YUV arrays to obtain an image array set; The arrangement unit is used to arrange the image array set according to preset horizontal and vertical coordinates to obtain the target image.

9. A DVR image processing device for reducing CPU load as claimed in claim 6, characterized in that: The encoding module includes a cropping unit and an encoding unit; The cropping unit is used to crop preset key information from the target image to obtain a key information tag set; wherein the preset key information includes date, time and gear position; The encoding unit is used to encode the key information tag set to generate the static image.

10. A DVR image processing device for reducing CPU load as claimed in claim 6, characterized in that: The merging module includes a merging unit, a key frame unit, a forward unit and a reverse unit; Wherein, the merging unit is used to merge the static image and the dynamic image in the target image to obtain a merged image; The key frame unit is used to add independent key frames to the combined image to obtain a final image; The forward unit is used to play the binocular view in the final image if the target vehicle is in a forward gear; The reversing unit is used to play the surround view in the final image if the target vehicle is in a reverse gear.

11. A storage medium, characterized in that: The storage medium stores a computer program, which is called and executed by a computer to implement a DVR image processing method for reducing CPU load as described in any one of claims 1 to 5 above.

Images