Image processing method and device

Abstract

The invention provides an image processing method and device, and relates to the technical field of vehicle-mounted image enhancement, and the method comprises the steps: obtaining an original image collected by an external camera in a vehicle; obtaining a distribution mode of interference information in the original image based on the original image; obtaining a target processing mode based on the distribution mode of the interference information; and processing the interference information in the original image based on the target processing mode to obtain a processed image, and displaying the processed image in a display screen of the vehicle. Based on the scheme, the image quality can be improved, and the driving safety is ensured.

Description

Technical Field

[0001] This application relates to the field of vehicle image enhancement technology, and in particular to an image processing method and apparatus. Background Technology

[0002] In off-road driving scenarios, especially in adverse weather conditions such as rain and snow, images captured by vehicle cameras often suffer from various forms of interference. This interference not only degrades image quality but also severely impairs the driver's ability to observe the vehicle's surroundings, posing a significant safety hazard to off-road driving.

[0003] Currently, fixed image processing methods are typically used to address various interference elements in an image. However, processing these interference elements with a fixed method results in poor image quality after interference removal, which fails to meet driving safety requirements.

[0004] Therefore, improving image quality and ensuring driving safety has become an urgent problem to be solved. Summary of the Invention

[0005] This application provides an image processing method and apparatus that can improve image quality and ensure driving safety.

[0006] Firstly, an image processing method is provided, the method comprising: Acquire raw images captured by external cameras inside and outside the vehicle; Based on the original image, the distribution pattern of interference information in the original image is obtained; Based on the distribution pattern of interference information, the target processing method is obtained; The interference information in the original image is processed based on the target processing method to obtain the processed image, which is then displayed on the vehicle's screen.

[0007] In the embodiments of this application, raw images captured by vehicle exterior cameras are acquired, and the distribution of interference information is analyzed, classifying the distribution of interference information into element-level and region-level distributions. Based on the distribution of interference information, a matching target processing method is further determined to achieve targeted processing of different interference features. Compared to existing technologies that use a single processing method for interference information in images, resulting in loss of detail or insufficient processing, this solution can determine the processing method for interference information based on its distribution. That is, different distributions of interference information can lead to different target processing methods, thereby achieving targeted processing of interference information in images, improving image quality, and ensuring driving safety.

[0008] In one implementation, the distribution method includes element-level distribution and / or region-level distribution.

[0009] It should be understood that element-level distribution refers to interference information acting on the smallest basic unit of an image (e.g., pixel, sub-pixel). The interference affects individual or discrete elements and does not form a continuous area coverage. The interference distribution exhibits the characteristics of being "dispersed and independent".

[0010] It should also be understood that regional distribution refers to interference information acting on continuous, patchy local areas in an image. The interference coverage includes multiple adjacent pixels, forming an "interference block" with a certain area and boundary. The interference distribution exhibits the characteristics of being "concentrated and continuous".

[0011] In conjunction with the first aspect, in certain implementations of the first aspect, based on the distribution of interference information, the target processing method is obtained, including: If the distribution of interference information is element-level, then filtering will be the target processing method. If the interference information is distributed at the regional level, then the repair process will be determined as the target processing method. If the distribution of interference information includes element-level distribution and region-level distribution, filtering and repair processing will be determined as the target processing methods.

[0012] In the embodiments of this application, the distribution pattern of interference information can reflect the presentation characteristics of interference in the image. Element-level distributed interference information usually exists in the form of discrete pixels in the image, with little impact on the overall image. Filtering can be identified as the target processing method to accurately remove interference information. Region-level distributed interference information, on the other hand, usually appears as continuously distributed pixel blocks in the image, with a greater impact on the overall image. If filtering is identified as the target processing method, it will destroy the overall image structure. Therefore, restoration processing can be identified as the target processing method. By identifying different target processing methods based on different distribution patterns of interference information, targeted processing of interference information can be performed using different methods, thereby improving image quality.

[0013] Combining the first aspect and the above implementation methods, in some implementation methods of the first aspect, the interference information in the original image is processed based on the target processing method to obtain the processed image, including: If the target processing method is filtering, the interference information distributed at the element level in the original image is filtered to obtain the processed image; If the target processing method is repair processing, the interference information distributed at the region level in the original image is repaired to obtain the processed image; If the target processing method includes filtering and repair processing, the interference information distributed at the element level in the original image is filtered, and the interference information distributed at the region level in the original image is repaired to obtain the processed image.

[0014] In the embodiments of this application, interference information in the original image is processed based on the target processing method. This allows for the processing of interference information in the original image according to different target processing methods, thereby achieving targeted processing of interference information and effectively improving image quality.

[0015] Combining the first aspect and the above implementation methods, in some implementation methods of the first aspect, interference information distributed at the region level in the original image is repaired, including: Image content detection is performed on the image regions where regionally distributed interference information is located, and the detection results are used to indicate whether the image content of the image region is complete. Based on the detection results, interference information distributed at the regional level in the original image is repaired.

[0016] In the embodiments of this application, content detection is performed on the image regions containing the regionally distributed interference information before the repair processing to obtain detection results; different repair processing is performed on the regionally distributed interference information in the original image according to different detection results. That is, the processing method for repair processing can be obtained from the detection results.

[0017] Combining the first aspect and the above implementation methods, in some implementation methods of the first aspect, based on the detection results, interference information distributed at the region level in the original image is repaired, including: If the detection result indicates that the image content is complete, the image display parameters of the image area are repaired, including brightness and / or contrast. If the detection result indicates that the image content is incomplete, the image content of the image area will be repaired.

[0018] In the embodiments of this application, targeted image repair processing is performed by distinguishing between two cases: complete image content and incomplete image content. When the image content is complete, image display parameters such as brightness and contrast are repaired, which can improve visual clarity while avoiding damage to the original texture and geometric details in the image. When the image content is incomplete, image content in the image area is repaired to restore the occluded or missing image content, thereby improving image quality.

[0019] Combining the first aspect and the above implementation methods, in some implementation methods of the first aspect, image display parameters of the image region are repaired, including: Based on the current terrain features of the vehicle and the terrain feature database, the target repair parameters corresponding to the current terrain features are obtained. The terrain feature database includes repair parameters corresponding to different terrain features. Based on the target repair parameters, the image display parameters of the image region are repaired.

[0020] In the embodiments of this application, based on the current terrain features where the vehicle is located, a correlation with the current terrain is obtained. Target restoration parameters based on feature matching. The target restoration parameters differ depending on the terrain features where the vehicle is located. Compared to existing technologies that use pre-configured fixed restoration parameters to restore image display parameters, this solution dynamically matches target restoration parameters based on the terrain features of the vehicle. This dynamic target restoration parameter improves the restoration effect of image display parameters and enhances image quality.

[0021] In conjunction with the first aspect and the above implementation methods, in some implementation methods of the first aspect, the image processing method further includes: Retrieve pre-saved image frames; Image content restoration processing is performed on the image region, including: Based on the image content of pre-saved image frames, the image content of the image region is repaired.

[0022] In one implementation, the pre-saved image frame includes the previous image frame of the original image.

[0023] In the embodiments of this application, the pre-saved image frames include the previous image frame of the original image. Since the original image and the previous image frame are two adjacent image frames, the image content of the two images has a high degree of similarity. When image region content is missing in the original image, the missing image content in the original image can be repaired using the previous image frame, ensuring the integrity of the image content in the original image and improving image quality.

[0024] In conjunction with the first aspect and the above implementation methods, in some implementation methods of the first aspect, the image processing method further includes: Based on the movement trajectory of the target object, identify the target camera that is obstructed in the vehicle; Save the image frames captured by the target camera.

[0025] In the embodiments of this application, by determining the obstructed target camera in the vehicle based on the movement trajectory of the target object, and saving the image frames captured by the target camera, it is possible to accurately identify the obstructed target camera and save only the image captured by that target camera. Compared to saving image frames from each camera, this solution can avoid redundant storage and reduce the ineffective use of storage resources.

[0026] Combining the first aspect and the above implementation methods, in some implementation methods of the first aspect, the distribution of interference information in the original image is obtained based on the original image, including: Target detection and image segmentation are performed on the original image to obtain the distribution pattern of interference information in the original image.

[0027] In the embodiments of this application, the distribution pattern of interference information in the original image is obtained by performing target detection and image segmentation on the original image. On the one hand, target detection can accurately identify and locate interference elements in the image that are distributed at the element level, achieving effective differentiation of multiple discrete interference elements in complex environments; on the other hand, image segmentation can segment interference regions in the image that are distributed at the region level. Therefore, based on target detection and image segmentation, the embodiments of this application can accurately identify various interference information in the original image.

[0028] In a second aspect, an image processing apparatus is provided, the apparatus comprising: The acquisition module is used to acquire raw images captured by the vehicle's external cameras; The processing module is used to obtain the distribution pattern of interference information in the original image based on the original image; obtain the target processing method based on the distribution pattern of interference information; process the interference information in the original image based on the target processing method to obtain the processed image, and display the processed image on the vehicle's display screen.

[0029] It should be understood that the extensions, limitations, explanations and descriptions of the relevant content in the first aspect above also apply to the same content in the second aspect.

[0030] Thirdly, a vehicle is provided, including a memory and a processor, the memory for storing executable program code, and the processor for calling and running the executable program code from the memory, causing the vehicle to perform the image processing method of the first aspect or any possible implementation thereof.

[0031] Fourthly, a computer program product is provided, comprising: computer program code, which, when run on a computer, causes the computer to perform the image processing method described in the first aspect or any possible implementation thereof.

[0032] Fifthly, a computer-readable storage medium is provided, which stores a computer program that, when executed, implements the image processing method described in the first aspect or any possible implementation thereof. Attached Figure Description

[0033] Figure 1 This is a schematic diagram illustrating an application scenario of the solution provided in an embodiment of this application; Figure 2 This is a schematic flowchart of an image processing method provided in an embodiment of this application; Figure 3 This is a schematic flowchart of another image processing method provided in an embodiment of this application; Figure 4 This is a schematic diagram of images before and after viewpoint correction provided in an embodiment of this application; Figure 5 This is a schematic diagram of an interface layout method provided in an embodiment of this application; Figure 6 This is a schematic diagram of the display interface of a display screen provided in an embodiment of this application; Figure 7 This is a schematic diagram of a vehicle system architecture provided in an embodiment of this application; Figure 8 This is a schematic diagram of the structure of an image processing apparatus provided in an embodiment of this application; Figure 9 This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application. Detailed Implementation

[0034] The technical solutions in this application will be clearly and thoroughly described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. "And / or" in the text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more than two.

[0035] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0036] Figure 1This is a schematic diagram illustrating an application scenario of the solution provided in an embodiment of this application.

[0037] like Figure 1 As shown, scenario 100 includes a vehicle 101 traveling on off-road terrain. Vehicle 101 includes multiple cameras, such as camera 102, camera 103, and camera 104.

[0038] Camera 102 is installed on the side of the vehicle body to collect environmental images around the vehicle 101; camera 103 and camera 104 are installed on the top of the vehicle to collect environmental images in front of the vehicle.

[0039] In one example, vehicle 101 is driving in rainy or snowy conditions. Cameras 102, 103, and 104 continuously capture images of the exterior of the vehicle. These images may include the outlines of trees to the side, the shape of the road ahead, obstacles around the vehicle, and various forms of interference information. For example, interference information may include discretely distributed raindrops or snowflakes, as well as continuous obstructions formed by splashing mud and water.

[0040] Currently, fixed image processing methods are typically used to address various interfering elements in an image. However, processing these interfering elements using a fixed method results in poor image quality after interference removal. Since drivers rely on in-cabin displays to observe their surroundings, poor image quality can impair their judgment of the environment, creating safety hazards during driving.

[0041] In view of this, this application provides an image processing method and apparatus. The image processing method provided by the embodiments of this application detects and analyzes the distribution of interference information in raw images captured by external cameras in a vehicle, classifying the distribution of interference information into element-level distribution and region-level distribution; based on different distribution patterns of interference information, a corresponding target processing method is determined; the target processing method is used to specifically process the interference information in the raw image, thereby obtaining a processed image, which is then displayed in real time on the vehicle's display screen.

[0042] The following is combined with Figures 2 to 7 The image processing method provided in the embodiments of this application will be described in detail.

[0043] Figure 2 This is a schematic flowchart of an image processing method provided in an embodiment of this application.

[0044] For example, Figure 2 The method 200 shown can be derived from Figure 1The execution can be performed by the vehicle 101 shown; or, it can be performed by the processor in the vehicle 101; or, it can be performed by the chip in the processor mounted in the vehicle 101.

[0045] like Figure 2 As shown, method 200 includes S201 to S204, which are described in detail below.

[0046] S201, acquire the raw images captured by the vehicle's exterior camera.

[0047] For example, the raw image is an unprocessed image of the external environment captured by an external camera. The environmental image includes road condition information, surrounding environment information, etc.

[0048] For example, exterior cameras include: front-view cameras, rear-view cameras, side-view cameras, and surround-view cameras.

[0049] In one implementation, the vehicle camera's acquisition mode is dynamically linked to the vehicle's driving status. That is, the vehicle camera can use different acquisition modes when the vehicle's driving status is different.

[0050] For example, when the vehicle is traveling at low speeds (e.g., speed less than or equal to 30 km / h), the camera activates panoramic acquisition mode, with the front, side, and rear cameras working simultaneously. The raw image includes environmental information around the vehicle (e.g., 360°). When the vehicle is traveling at high speeds (e.g., speed greater than 60 km / h), the system automatically switches to single-view front acquisition mode, and the side and rear cameras enter sleep mode to reduce vehicle power consumption. In this case, the raw image only contains road information within the forward (e.g., 120°) field of view.

[0051] Optionally, the external camera continuously collects environmental information outside the vehicle at a preset frame rate.

[0052] Furthermore, after acquiring the raw images, they can be saved. For example, the raw images can be transmitted to the vehicle's image buffer via the vehicle bus. The buffer can employ a first-in, first-out (FIFO) mechanism to ensure that the most recently acquired image data is prioritized for subsequent processing steps.

[0053] Optionally, the above image processing method further includes: Based on the movement trajectory of the target object, identify the target camera that is obstructed in the vehicle; Save the image frames captured by the target camera.

[0054] For example, when saving the original images, the original images captured by each of the vehicle's external cameras can be saved. Alternatively, if an obstructing target camera is identified among the external cameras, the original image captured by that target camera can be saved. In this case, the obstructed target camera in the vehicle is identified based on the movement trajectory of the target object.

[0055] For example, based on the current vehicle's operating status information and tire tread parameters, a rain and snow splash model is constructed to dynamically simulate the mud and water splashing process, obtain the motion trajectory of the mud and water splash, and based on the motion trajectory, identify the target camera in the vehicle that may be obstructed, and save the original image collected when the camera is not obstructed.

[0056] It should be noted that tire tread parameters include at least one of the following: groove depth, angle, and distribution density.

[0057] S202, Based on the original image, obtain the distribution pattern of interference information in the original image.

[0058] Optionally, target detection and image segmentation can be performed on the original image to obtain the distribution pattern of interference information in the original image.

[0059] It should be understood that element-level distribution refers to interference information acting on the smallest basic unit of an image (e.g., pixel, sub-pixel). The interference affects individual or discrete elements and does not form a continuous area coverage. The interference distribution exhibits the characteristics of being "dispersed and independent".

[0060] For example, the snowflakes and raindrops in the original image are distributed at the element level, and these disturbances exist in the form of scattered dots or tiny patches.

[0061] It should also be understood that regional distribution refers to interference information acting on continuous, patchy local areas in an image. The interference coverage includes multiple adjacent pixels, forming an "interference block" with a certain area and boundary. The interference distribution exhibits the characteristics of being "concentrated and continuous".

[0062] For example, the distribution of rain curtains formed by heavy rain and large areas of mud and water splashed up by vehicles in the original image is regional, and these disturbances occupy a certain complete area of ​​the image.

[0063] In one implementation, the distribution pattern of interference information in the original image is obtained based on the original image, including: performing target detection and image segmentation on the original image to obtain the distribution pattern of interference information in the original image.

[0064] Furthermore, in vehicle scenarios, the raw images captured by external cameras may exhibit geometric distortion due to factors such as installation location and lens characteristics. Directly using these images for interference identification could lead to misjudgments of the interference's location and extent. Therefore, it is advisable to first correct the viewing angle of the raw images before performing interference identification.

[0065] In another implementation, based on the original image, the distribution pattern of interference information in the original image is obtained, including: performing viewpoint correction on the original image to obtain a corrected image; and performing target detection and image segmentation on the corrected image to obtain the distribution pattern of interference information in the original image.

[0066] Optionally, perspective correction and compensation parameters are obtained based on the vehicle's driving status information; perspective correction is performed on the original image based on the perspective correction and compensation parameters to obtain the corrected image; wherein, the driving status information includes at least one of the following: vehicle body posture change, real-time vehicle speed, and vehicle steering angle; perspective correction and compensation parameters include horizontal offset compensation and vertical offset compensation.

[0067] For example, the vehicle's attitude change can be obtained through a three-dimensional accelerometer and gyroscope, the real-time speed of the vehicle can be obtained through a wheel speed sensor, and the steering angle of the vehicle can be obtained through a steering angle sensor.

[0068] In one example, after obtaining the vehicle's driving status information, the horizontal offset compensation amount is obtained using the following formula: ; in, This indicates the amount of horizontal offset compensation. Indicates the vehicle's speed; Indicates the vehicle's steering angle; Indicates the sampling time interval.

[0069] For example, when a vehicle turns left, its speed... The speed is 10 m / s, and the turning angle is... 30°, sampling time interval If it is 0.5 seconds, then the horizontal offset compensation amount is... The distance is 6.25 meters. When the conversion relationship is 10 pixels / meter, the corresponding image pixel compensation is 62.5 pixels. That is, the image needs to be compensated 62.5 pixels to the left in the horizontal direction to offset the rightward deviation of the image caused by the turning.

[0070] In another example, after obtaining the vehicle's driving status information, the vertical offset compensation amount is obtained using the following formula: ; in, This indicates the vertical offset compensation amount; Indicates the installation height of the vehicle camera; Indicates the vehicle's steering angle; This represents the conversion factor between pixels and actual physical distance.

[0071] For example, when the vehicle accelerates rapidly, the sensor detects the pitch angle. The camera installation height is 2°. It is 1.6 meters, conversion factor If the value is 50 pixels per meter, then the vertical offset compensation... The vertical compensation is approximately 2.79 pixels, meaning the image needs to be vertically compensated downwards by 2.79 pixels to correct the upward shift of the image caused by the car's front being tilted upwards.

[0072] For example, Figure 4 This is a schematic diagram of images before and after viewpoint correction provided in an embodiment of this application.

[0073] like Figure 4 As shown, it includes: image 401 and image 402. Among them, image 401 is the original image, and image 402 is the corrected image.

[0074] For example, when a vehicle is driving on an off-road slope, the camera tilts with the vehicle, resulting in distortion of the road and scenery; road markings and trees are tilted to the left. Furthermore, the spatial relationships of the road in the image are distorted due to the vehicle's changing posture. In this situation, the vehicle's driving status information obtained from sensors, combined with the camera's installation height, is used to calculate perspective correction compensation parameters, and the original image is then corrected for perspective.

[0075] In this embodiment, by using perspective correction processing, the impact of vehicle posture changes on camera imaging angle can be effectively reduced, so that the corrected image can more accurately reflect the actual environmental conditions around the vehicle, providing higher quality image input for subsequent processing.

[0076] Optionally, after obtaining the viewpoint-corrected image through the above method, the corrected image is subjected to deep learning-based target detection and dynamic threshold-based image segmentation to obtain the distribution pattern of interference information in the original image.

[0077] For example, the YOLOv5s deep learning object detection algorithm can be used to identify interfering elements. Since the interfering elements are small in scale, small-scale interfering samples can be added and the anchor box size optimized during model training. The training set contains 120,000 annotated images of off-road rain and snow scenes, with annotation types covering flying snowflakes, splashing mud, etc. During inference, the original images are input into the trained model, and detection results with high confidence (e.g., 0.85) are selected. Duplicate annotations are removed, and the final output is a set of coordinates for element-level distributed interfering information, thus achieving the labeling and classification of interfering elements.

[0078] For example, a dynamic thresholding algorithm based on color space analysis can be used to segment the original image and identify interference regions. For instance, the image can be converted to a specific hue-saturation-luminance color space, a local dynamic threshold can be calculated using an adaptive block-segmentation strategy, and binarization segmentation can be performed to obtain a mask of the interference region. Morphological operations can then be used to optimize the mask contour. The adaptive block-segmentation strategy refers to dividing the image pixels into several sub-blocks based on the image pixel distribution density characteristics, and generating a specific local threshold for each sub-block by statistically analyzing its pixel value distribution.

[0079] It should be understood that the interference information in the image includes interference elements and interference regions. The interference elements are distributed at the element level, while the interference regions are distributed at the region level.

[0080] In the embodiments of this application, the distribution pattern of interference information in the original image is obtained by performing target detection and image segmentation on the original image. On the one hand, target detection can accurately identify and locate interference elements in the image that are distributed at the element level, achieving effective differentiation of multiple discrete interference elements in complex environments; on the other hand, image segmentation can segment interference regions in the image that are distributed at the region level. Therefore, based on target detection and image segmentation, the embodiments of this application can accurately identify various interference information in the original image.

[0081] S203, based on the distribution of interference information, the target processing method is obtained.

[0082] It should be understood that the target processing method for interference information can be determined based on the distribution pattern of the interference information in the image. This means that different distribution patterns of interference information will result in different target processing methods; thus, targeted processing of interference information in the image can be achieved, thereby improving image quality.

[0083] Optionally, based on the distribution pattern of the interference information, the target processing method is obtained, including: if the distribution pattern of the interference information is element-level distribution, filtering processing is determined as the target processing method; if the distribution pattern of the interference information is region-level distribution, repair processing is determined as the target processing method; if the distribution pattern of the interference information includes both element-level and region-level distribution, filtering processing and repair processing are determined as the target processing methods.

[0084] It should be understood that element-wise distributed interference information typically exists as discrete pixels in an image and has a relatively small impact on the overall image. Therefore, filtering is used to process element-wise distributed interference information, thereby accurately removing this interference information without affecting other normal areas in the image.

[0085] It should also be understood that regionally distributed interference information typically appears as continuously distributed pixel blocks in an image, significantly impacting the overall image. Filtering this regionally distributed interference information may disrupt the overall image structure. Therefore, inpainting techniques can be used to process regionally distributed interference information, ensuring visual consistency between the affected area and its surrounding regions, thereby improving image quality.

[0086] For example, filtering refers to processing the pixel values ​​at the location of interference information based on pixel-level or local region features to suppress or remove the interference information. Filtering can include any one of smoothing, mean filtering, median filtering, high-pass filtering, or low-pass filtering.

[0087] For example, repair processing refers to using technical means to repair and restore areas in an image that contain interfering information, or to repair parameters in the image area where the interfering information is located.

[0088] In the embodiments of this application, the distribution pattern of interference information can reflect the presentation characteristics of interference in the image. Element-level distributed interference information usually exists in the form of discrete pixels in the image, with little impact on the overall image. Filtering can be identified as the target processing method to accurately remove interference information. Region-level distributed interference information, on the other hand, usually appears as continuously distributed pixel blocks in the image, with a greater impact on the overall image. If filtering is identified as the target processing method, it will destroy the overall image structure. Therefore, restoration processing can be identified as the target processing method. By identifying different target processing methods based on different distribution patterns of interference information, targeted processing of interference information can be performed using different methods, thereby improving image quality.

[0089] S204: Based on the target processing method, the interference information in the original image is processed to obtain the processed image, and the processed image is displayed on the vehicle's display screen.

[0090] Optionally, the interference information in the original image is processed based on the target processing method to obtain a processed image, including: If the target processing method is filtering, the interference information distributed at the element level in the original image is filtered to obtain the processed image; if the target processing method is repair, the interference information distributed at the region level in the original image is repaired to obtain the processed image; if the target processing method includes both filtering and repair, the interference information distributed at the element level in the original image is filtered, and the interference information distributed at the region level in the original image is repaired to obtain the processed image.

[0091] In the embodiments of this application, interference information in the original image is processed based on the target processing method. This allows for the processing of interference information in the original image according to different target processing methods, thereby achieving targeted processing of interference information and effectively improving image quality.

[0092] The implementation methods of filtering and repair processing are described in detail below.

[0093] Optionally, interference information distributed at the region level in the original image is repaired, including: Image content detection is performed on the image regions where regionally distributed interference information is located, and the detection results are used to indicate whether the image content of the image region is complete. Based on the detection results, the regionally distributed interference information in the original image is repaired.

[0094] For example, when a vehicle is driving in heavy snow, dense snowflakes form large, blurred white areas in the image, which may cover the stop line at the intersection ahead. In this case, a region feature extraction algorithm is used to scan the image area obscured by the snowflakes, analyzing the pixel continuity, edge integrity, and feature correlation with surrounding areas. If a broken stop line edge or road surface texture is detected within the obscured area, the image is determined to be incomplete; if the snowflakes obscure a sky area without crucial information, the image is determined to be complete.

[0095] In the embodiments of this application, content detection is performed on the image regions containing the regionally distributed interference information before the repair processing to obtain detection results; different repair processing is performed on the regionally distributed interference information in the original image according to different detection results. That is, the processing method for repair processing can be obtained from the detection results.

[0096] Optionally, based on the detection results, the interference information distributed at the regional level in the original image is repaired, including: if the detection results indicate that the image content is complete, the image display parameters of the image region are repaired, including brightness and / or contrast; if the detection results indicate that the image content is incomplete, the image content of the image region is repaired.

[0097] In the embodiments of this application, image quality is improved by using different processing methods to specifically process interference information in the image.

[0098] In the embodiments of this application, targeted image repair processing is performed by distinguishing between two cases: complete image content and incomplete image content. When the image content is complete, image display parameters such as brightness and contrast are repaired, which can improve visual clarity while avoiding damage to the original texture and geometric details in the image. When the image content is incomplete, image content in the image area is repaired to restore the occluded or missing image content, thereby improving image quality.

[0099] Optionally, image display parameters of the image region are repaired, including: Based on the current terrain features of the vehicle and the terrain feature database, the target repair parameters corresponding to the current terrain features are obtained. The terrain feature database includes repair parameters corresponding to different terrain features. Based on the target repair parameters, the image display parameters of the image area are repaired.

[0100] For example, the terrain feature database includes: landform templates and corresponding repair parameters for each landform template. When repairing image display parameters, the current terrain feature is matched with the terrain feature database to obtain the target repair parameters corresponding to the current terrain. Based on the obtained target parameters, the display parameters of the image area are repaired.

[0101] In one example, environmental data collected by sensors in the vehicle is acquired; based on the environmental data collected by the vehicle sensors, the current terrain features where the vehicle is located are obtained. The environmental data includes any one of the following: visual environmental data, physical environmental data, and vehicle status and interaction environmental data.

[0102] For example, visual environmental data includes LiDAR point cloud data; physical environmental data includes millimeter-wave radar data and ultrasonic radar data; and vehicle status and interaction environmental data includes wheel speed sensor data.

[0103] Optionally, based on the current environmental data acquired by the vehicle sensors, the current terrain features where the vehicle is located are obtained; based on the current terrain features where the vehicle is located and the terrain feature database, the target repair parameters corresponding to the current terrain features are obtained.

[0104] For example, the vehicle obtains the current terrain features based on the current environmental data acquired by sensors, and matches these features with a terrain feature database to obtain the corresponding repair parameters. For instance, after matching the current terrain features with the database, it is determined that the vehicle is currently on a snow-covered road surface, and the terrain feature database records the repair parameters corresponding to the snow-covered road surface.

[0105] For example, repair parameters refer to key parameters used to adjust the image display effect, including: brightness parameters, contrast parameters, saturation parameters, white balance parameters, etc.

[0106] In the embodiments of this application, target restoration parameters matching the current terrain features are obtained based on the current terrain features where the vehicle is located. The obtained target restoration parameters differ depending on the terrain features where the vehicle is located. Compared to the prior art, which uses pre-configured fixed restoration parameters to restore image display parameters of an image region, this solution enables dynamic matching of target restoration parameters based on the terrain features where the vehicle is located. Through these dynamic target restoration parameters, the restoration effect of image display parameters can be improved, thereby enhancing image quality.

[0107] Optionally, the above image processing method further includes: Retrieve pre-saved image frames; Image content restoration processing is performed on the image region, including: Based on the image content of pre-saved image frames, the image content of the image region is repaired.

[0108] For example, when a vehicle travels through a tree-shaded section of road, the road markings 5 ​​meters ahead are obscured by tree branches in the original image captured by the camera. In this case, a pre-saved image frame of the same road section without obstruction is retrieved first. This frame contains the complete road markings. Then, based on the shape, color, and positional features of the road markings in the pre-saved image frame, the pixel information of the corresponding markings in the pre-saved image frame is filled into the obscured area, thus completing the image content restoration.

[0109] Furthermore, after repairing the image content of the image area, it can be determined whether the image display parameters need to be repaired again based on the current environment.

[0110] Furthermore, real-time environmental parameters are acquired, and based on these parameters, the image display parameters are determined to be repaired.

[0111] In one implementation, the pre-saved image frame includes the previous image frame of the original image.

[0112] In the embodiments of this application, the pre-saved image frames include the previous image frame of the original image. Since the original image and the previous image frame are two adjacent image frames, the image content of the two images has a high degree of similarity. When image region content is missing in the original image, the missing image content in the original image can be repaired using the previous image frame, ensuring the integrity of the image content in the original image and improving image quality.

[0113] Furthermore, after obtaining the processed image, the processed image is displayed on the vehicle's screen.

[0114] In one implementation, the layout of the display screen is determined based on the vehicle's current driving mode.

[0115] For example, in off-road rain and snow mode, the display uses a three-section layout. The left 30% area displays the original image, that is, the unprocessed raw image captured by the camera in real time; the right 60% area displays the enhanced image, that is, the image after view correction, interference filtering, and region restoration; and the bottom 10% area displays warning information.

[0116] For example, the layout of the display screen in three sections is as follows: Figure 5 As shown. Figure 5 The layout shown includes: a display area 501 for the original image, a display area 502 for the enhanced image, and a display area 503 for the warning information.

[0117] The system includes three main areas: 501 for displaying the original image (captured by the vehicle's camera without processing), 502 for displaying the enhanced image (captured through target processing), and 503 for displaying warning information about the vehicle's current location (e.g., obstacle warnings, road condition alerts). The area 501 displays warning information about the vehicle's location, such as obstacle warnings and road condition alerts, providing the driver with additional information.

[0118] Optionally, when the driving mode is Eco mode, the display screen is single-screen; when the driving mode is Sport mode, the display screen is two-part layout.

[0119] It should be understood that the above are examples illustrating the display layout for different driving modes; this application does not limit the scope of the embodiments.

[0120] For example, when using a three-section layout, the display interface of the screen is as follows: Figure 6 As shown.

[0121] The display interface 600 includes: current warning information around the vehicle, and two sub-areas for displaying the road scene. The left sub-area displays the raw image captured by the vehicle's camera without processing, and the right sub-area displays the image after perspective correction and enhancement processing.

[0122] For example, a driver can select a key monitoring area using touch operation. If the driver selects a key monitoring area in the cabin, the display interface will only show the image content of the key monitoring area selected by the driver. For instance, if the driver selects the area around the wheels using touch operation, the display screen in the cabin will only show the original image of the area around the wheels, the enhanced image, and warning information about the vehicle's current position.

[0123] In the embodiments of this application, interference information is detected and its distribution analyzed in the raw images captured by the vehicle's external cameras. The interference information is categorized into element-level and region-level distributions. Based on the different distribution patterns of the interference information, corresponding target processing methods are determined. These target processing methods are then used to specifically process the interference information in the raw images, resulting in a processed image, which is then displayed in real-time on the vehicle's screen. This solution improves image quality and ensures driving safety by specifically processing interference information with different distribution patterns.

[0124] Figure 3 This is a schematic flowchart of an image processing method provided in an embodiment of this application.

[0125] Figure 3 The method 300 shown can be derived from Figure 1 The execution can be performed by the vehicle 101 shown; or, it can be performed by the processor in the vehicle 101; or, it can be performed by the chip in the processor mounted in the vehicle 101.

[0126] like Figure 3 As shown, method 300 includes S301 to S309, which are described in detail below.

[0127] S301, acquire the raw image captured by the vehicle's camera.

[0128] For example, after the vehicle is started, the vehicle's front-view camera and side-view camera continuously collect raw images of the outside of the vehicle at a preset frame rate.

[0129] Alternatively, the implementation of S301 can be found in [reference needed]. Figure 2 The relevant descriptions of S201 will not be repeated here.

[0130] S302, perform perspective correction on the acquired original image to obtain the perspective-corrected image.

[0131] For example, based on the vehicle's driving status information, perspective correction and compensation parameters are obtained; based on the perspective correction and compensation parameters, the original image is corrected to obtain a perspective-corrected image.

[0132] Alternatively, the implementation of S302 can be found in [reference needed]. Figure 2The relevant description of S202 will not be repeated here.

[0133] S303 uses a target detection algorithm on the corrected image to identify whether there are interfering elements in the image.

[0134] For example, a deep learning-based object detection algorithm is used on the corrected image to identify whether there are interfering elements in the image. If there are interfering elements, then S305 is executed.

[0135] It is understandable that the model in this embodiment balances detection accuracy and inference speed, and can meet the needs of real-time image processing during vehicle movement.

[0136] Alternatively, the implementation of S303 can be found in [reference needed]. Figure 2 The relevant description of S202 will not be repeated here.

[0137] S304. Apply a dynamic threshold segmentation algorithm to the corrected image to identify whether there are interference regions in the image.

[0138] For example, for the image after viewpoint correction, a dynamic threshold segmentation algorithm based on color space analysis is used to identify whether there are interference regions in the image. If there are interference regions, then S306 is executed.

[0139] Alternatively, the implementation of S304 can be found in [reference needed]. Figure 2 The relevant description of S202 will not be repeated here.

[0140] S305. If there are interfering elements, filter them to obtain the enhanced image.

[0141] For example, if there are interfering elements in the corrected image, the interfering elements are filtered out to obtain the processed image.

[0142] Alternatively, the implementation of S305 can be found in [reference needed]. Figure 2 The relevant descriptions of S203 and S204 will not be repeated here.

[0143] S306. If there is an interference area, determine whether the image content corresponding to the interference area is complete; if yes, proceed to 307; if no, proceed to 308.

[0144] For example, image content detection is performed on the image region where the interference information is located to obtain the detection result, which is used to indicate whether the image content of the image region is complete; based on the detection result, the interference information distributed at the region level in the original image is repaired.

[0145] It should be understood that for scenarios with limited computing power, the judgment logic can be simplified. For example, by calculating the texture entropy value of the interference area, if the difference between the texture entropy of the area and the surrounding normal area is less than 0.3, the content is judged to be complete; otherwise, the content is judged to be incomplete.

[0146] Alternatively, the implementation of S306 can be found in [reference needed]. Figure 2 The relevant descriptions in S204 will not be repeated here.

[0147] S307. Based on the geomorphological feature database, the parameters of the segmented interference areas are repaired to obtain the enhanced image.

[0148] For example, based on the current terrain features where the vehicle is located and the terrain feature database, the target repair parameters corresponding to the current terrain features are obtained. The terrain feature database includes repair parameters corresponding to different terrain features. Based on the target repair parameters, the image display parameters of the image area are repaired.

[0149] It should be understood that the geomorphological feature database includes geomorphological template data for various typical off-road scenarios. For each geomorphological type, the database records the corresponding repair parameters, including brightness and contrast.

[0150] Alternatively, the implementation of S307 can be found in [reference needed]. Figure 2 The relevant descriptions in S204 will not be repeated here.

[0151] S308. Based on the pre-stored previous frame image, perform content restoration on the image corresponding to the segmented interference region to obtain the enhanced image.

[0152] Understandably, the system can pre-screen the stored previous frame image, retaining only images with acceptable sharpness and no significant motion blur as reference frames. During the restoration process, if there are differences in lighting conditions between the previous and current frames, the corresponding areas of the reference image are first corrected for brightness and color temperature before restoration.

[0153] Alternatively, the implementation of S308 can be found in [reference needed]. Figure 2 The relevant descriptions in S204 will not be repeated here.

[0154] S309. Display the enhanced image.

[0155] For example, the enhanced image obtained through the above processing is output to the vehicle display terminal for display.

[0156] It is understood that in-vehicle display terminals include, but are not limited to, central control displays, head-up displays, or in-vehicle augmented reality display devices.

[0157] Alternatively, the implementation of S309 can be found in [reference needed]. Figure 2 The relevant descriptions in S204 will not be repeated here.

[0158] In the embodiments of this application, the distribution pattern of interference information can reflect the presentation characteristics of interference in the image. Element-level distribution usually exists in the form of discrete pixels in the image, with little impact on the overall image. Filtering can be identified as the target processing method to accurately remove interference information. Region-level distribution interference information, on the other hand, usually appears as continuously distributed pixel blocks in the image, with a greater impact on the overall image. If filtering is identified as the target processing method, it will destroy the overall image structure. Therefore, restoration processing can be identified as the target processing method. By determining different target processing methods based on different distribution patterns of interference information, different processing methods can be used to specifically process the interference information and improve image quality.

[0159] Optionally, in one implementation, the environmental information of the current vehicle is obtained, and based on the obtained environmental information, an image processing method for removing interference information from the image is selected. The environmental information includes current weather information; the image processing method includes the image processing method described in this embodiment or a pre-configured image processing method.

[0160] For example, if the environmental information indicates that the current weather is sunny, the image captured by the camera will be relatively clear. In this case, the interference information in the original image can be processed using the pre-configured filtering process or the fixed processing method in the pre-configured repair process. If the current weather information indicates that the current weather is rainy or snowy, the image captured by the camera will have relatively more interference information, and the interference information may include multiple distribution patterns. In this case, the image processing method in this embodiment of the application will be used to process the interference information in the original image.

[0161] Figure 7 This is a schematic diagram of a vehicle system architecture provided in an embodiment of this application. Figure 7 As shown, the vehicle's system architecture includes a camera 701, a three-dimensional acceleration sensor 702, a gyroscope 703, a vehicle speed and rotational speed sensor 704, a steering angle sensor 705, an on-board host 708, and a display screen 714.

[0162] Camera 701 can be used to capture raw images of the vehicle's surroundings, including the road environment and obstacles, providing raw images for subsequent processing and display.

[0163] Alternatively, camera 701 may refer to Figure 1 Any one of cameras 102, 103, and 101.

[0164] The 3D accelerometer 702 can be used to detect the acceleration information of a vehicle, determine the vehicle's motion state and driving conditions, and provide reference data for driving scene recognition and image correction.

[0165] The gyroscope 703 can be used to detect the vehicle's attitude information, including pitch, roll and yaw angles, to help determine the vehicle's current attitude state.

[0166] The steering angle sensor 704 can be used to detect the rotation angle of the vehicle's steering wheels and determine the steering amplitude of the vehicle.

[0167] The display screen 714 can be used to display the original image captured by the camera, the enhanced image, and warning information about the current environment.

[0168] Optionally, the display screen 714 can be used to perform Figure 2 The processed image is displayed in S204; or, the display screen 714 can be used to perform... Figure 3 The enhanced image is displayed in S309. The processed image in S204 can refer to the enhanced image in S309.

[0169] The vehicle-mounted host 708, also known as a vehicle infotainment system, host, vehicle-mounted host system, cockpit domain controller, vehicle information processing unit, multimedia host, cockpit host, terminal information display unit, etc., is the core control unit of the vehicle's electronic system. It is responsible for receiving data from various sensors and cameras, and performing functions such as sensor data acquisition, camera control, interference information processing, human-machine interaction and display control. The vehicle-mounted host 708 may include a data fusion processing module 706, a rain and snow processing module 707, a dynamic interference layering module 709, and a human-machine interaction enhancement processing module 713.

[0170] The data fusion processing module 706 is used to fuse multi-source data from a three-dimensional accelerometer, gyroscope, wheel speed sensor, and steering angle sensor to accurately locate the vehicle's spatial position and attitude, thereby assisting in image correction and interference information processing.

[0171] Understandably, the data fusion processing module 706 can be used to perform... Figure 3 In step S302, the viewpoint of the acquired raw image is corrected.

[0172] The rain and snow processing module 707 is used to combine the three-dimensional morphological parameters of the tire tread and the real-time operating status of the vehicle to dynamically simulate the mud and water splashing process, predict the time and location range that may obstruct the camera, and collect and store the image frames of the area in the unobstructed state in advance, which are used as the reference image of the previous frame for image content repair.

[0173] The human-computer interaction enhancement processing module 713 is used to realize the split-screen display function, which divides the display interface into the original image display area, the enhanced image display area, and the warning information display area. This allows the driver to efficiently obtain multiple information on the same interface, promptly remind the driver to avoid risks, and effectively improve driving safety and reliability.

[0174] The dynamic interference layering module 709 is used to identify interference elements in the image and segment the regions containing interference. By layering different types of interference, appropriate processing methods can be adopted for different interference features, thereby improving the processing efficiency of interference information. The dynamic interference layering module 709 includes a dual-channel parallel processing module 710, a terrain feature database 711, and a regional repair module 712.

[0175] The dual-channel parallel processing module 710 is used to perform deep learning-based interference element target detection and dynamic threshold segmentation interference region segmentation on the corrected image. The deep learning-based target detection algorithm is used to identify interference elements, and the dynamic threshold segmentation interference region segmentation is used to extract large-scale continuous interference regions. The dual-channel parallel processing module can efficiently identify discrete interference elements and segment large-scale interference regions.

[0176] Optionally, the dual-channel parallel processing module 710 can be used to execute... Figure 2 In step S202, the distribution pattern of interference information in the original image is obtained based on the original image; alternatively, the dual-channel parallel processing module 710 can be used to perform... Figure 3 In step S303, interference elements in the image are identified, and in step S304, interference regions in the image are segmented.

[0177] The terrain feature database 711 is used to store terrain templates for various typical off-road scenarios (such as snow-covered roads, muddy puddles, etc.) and record image feature parameters such as brightness and contrast for each terrain under different lighting and weather conditions. This is used for terrain category matching and calling the corresponding repair parameters during the processing.

[0178] The regional repair module 712 is used to call the parameter repair strategy corresponding to the landform category based on the matching results of the terrain feature database, optimize and adjust parameters such as brightness and contrast in the interference area, and maintain the continuity with the surrounding area in terms of geometric structure and texture details.

[0179] Optionally, the regional repair module 712 is also used to repair the image content of an image region based on the image content of a pre-saved image frame.

[0180] It should be understood that the above examples are provided to help those skilled in the art understand the embodiments of this application, and are not intended to limit the embodiments of this application to the specific values ​​or scenarios illustrated. Those skilled in the art can obviously make various equivalent modifications or changes based on the above examples, and such modifications or changes also fall within the scope of the embodiments of this application.

[0181] The above text combined Figures 1 to 7 The image processing method provided in the embodiments of this application has been described in detail; the following will be combined with Figures 8 to 9 The embodiments of the apparatus provided in this application are described in detail below. It should be understood that the apparatus in the embodiments of this application can execute the various methods of the foregoing embodiments of this application, that is, the specific working processes of the various products below can be referred to the corresponding processes in the foregoing method embodiments.

[0182] Figure 8 This is a schematic diagram of the structure of an image processing apparatus provided in an embodiment of this application.

[0183] For example, such as Figure 8 As shown, the image processing apparatus 800 includes: The acquisition module 801 is used to acquire raw images captured by the vehicle's external camera; The processing module 802 is used to obtain the distribution pattern of interference information in the original image based on the original image; obtain the target processing method based on the distribution pattern of interference information; process the interference information in the original image based on the target processing method to obtain the processed image, and display the processed image on the vehicle's display screen.

[0184] Optionally, as an embodiment, the processing module 802 is specifically used for: If the interference information is distributed at the element level, filtering is determined as the target processing method; if the interference information is distributed at the region level, repair is determined as the target processing method; if the interference information is distributed at both the element level and the region level, filtering and repair are determined as the target processing methods.

[0185] Optionally, as an embodiment, the processing module 802 is specifically used for: If the target processing method is filtering, the interference information distributed at the element level in the original image is filtered to obtain the processed image; if the target processing method is repair, the interference information distributed at the region level in the original image is repaired to obtain the processed image; if the target processing method includes both filtering and repair, the interference information distributed at the element level in the original image is filtered, and the interference information distributed at the region level in the original image is repaired to obtain the processed image.

[0186] Optionally, as an embodiment, the processing module 802 is specifically used for: Image content detection is performed on the image regions where regionally distributed interference information is located, and the detection results are used to indicate whether the image content of the image region is complete. Based on the detection results, the regionally distributed interference information in the original image is repaired.

[0187] Optionally, as an embodiment, the processing module 802 is specifically used for: If the detection result indicates that the image content is complete, the image display parameters of the image area are repaired, including brightness and / or contrast; if the detection result indicates that the image content is incomplete, the image content of the image area is repaired.

[0188] Optionally, as an embodiment, the processing module 802 is specifically used for: Based on the current terrain features of the vehicle and the terrain feature database, the target repair parameters corresponding to the current terrain features are obtained. The terrain feature database includes repair parameters corresponding to different terrain features. Based on the target repair parameters, the image display parameters of the image area are repaired.

[0189] Optionally, as an embodiment, the processing module 802 is further configured to: Acquire pre-saved image frames; perform image content restoration processing on the image region, including: performing image content restoration processing on the image region based on the image content of the pre-saved image frames.

[0190] Optionally, as an embodiment, the processing module 802 is further configured to: Based on the motion trajectory of the target object, the target camera in the vehicle that is obstructed is identified; the image frames captured by the target camera are saved.

[0191] Optionally, as an embodiment, the processing module 802 is specifically used for: Target detection and image segmentation are performed on the original image to obtain the distribution pattern of interference information in the original image.

[0192] It should be noted that the image processing apparatus 800 described above is embodied in the form of a functional unit. The term "module" here can be implemented in software and / or hardware, without specific limitations.

[0193] For example, a "module" can be a software program, a hardware circuit, or a combination of both that implements the above functions. The hardware circuit may include an application-specific integrated circuit (ASIC), electronic circuits, a processor (e.g., a shared processor, a proprietary processor, or a group processor) and memory for executing one or more software or firmware programs, integrated logic circuits, and / or other suitable components that support the described functions.

[0194] Therefore, the units of the various examples described in the embodiments of 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.

[0195] Figure 9 This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application.

[0196] For example, vehicle 901 includes: processor 902, memory 903 and executable program code 904.

[0197] For example, vehicle 901 includes one or more processors 902, which can support vehicle 901 in implementing the image processing method in the method embodiment. Processor 902 can be a general-purpose processor or a special-purpose processor. For example, processor 902 can be a central processing unit (CPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, such as discrete gates, transistor logic devices, or discrete hardware components.

[0198] For example, the processor 902 can be used to control the vehicle 901, execute software programs, and process data from the software programs. The vehicle 901 may also include a communication unit for receiving and transmitting signals.

[0199] For example, the vehicle 901 may include one or more memories 903 storing executable program code 904. The executable program code 904 can be run by the processor 902 to generate instructions, causing the processor 902 to execute the image processing method described in the above method embodiments according to the instructions. For example, the processor 902 executes the following according to the instructions: acquiring the original image captured by the external camera in the vehicle; obtaining the distribution pattern of interference information in the original image based on the original image; obtaining the target processing method based on the distribution pattern of interference information; processing the interference information in the original image based on the target processing method to obtain the processed image, and displaying the processed image on the vehicle's display screen.

[0200] Optionally, the memory 903 may also store data. Optionally, the processor 902 may also read data stored in the memory 903, which may be stored at the same memory address as the executable program code 904, or the data may be stored at a different memory address than the executable program code 904.

[0201] For example, the processor 902 and memory 903 can be configured separately or integrated together, for example, integrated on the system-on-chip (SOC) of the terminal device.

[0202] For example, memory 903 can be used to store related programs of the image processing method provided in the embodiments of this application, and processor 902 can be used to call the executable program code 904 stored in memory 903 when controlling a vehicle to execute the image processing method of the embodiments of this application. This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the image processing method of any of the foregoing embodiments.

[0203] The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, Digital Video Discs (DVDs), Compact Disc Read-Only Memory (CD-ROMs), microdrives, and magneto-optical disks, read-only memory (ROMs), random access memory (RAMs), erasable programmable read-only memory (EPROMs), electrically erasable programmable read-only memory (EEPROMs), dynamic random access memory (DRAMs), video random access memory (VRAMs), flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0204] This application also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to implement the image processing method described in the above embodiments.

[0205] In addition, the vehicle provided in the embodiments of this application may specifically be a chip, component or module. The vehicle may include a connected processor and a memory. The memory is used to store instructions. When the vehicle is running, the processor may call and execute the instructions to make the chip perform the image processing method in the above embodiments.

[0206] The vehicle, computer-readable storage medium, computer program product or chip provided in this application are all used to execute the corresponding image processing method provided above. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding image processing method provided above, and will not be repeated here.

[0207] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0208] 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 modules or 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 device, 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 devices or units may be electrical, mechanical, or other forms.

[0209] 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. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An image processing method, characterized in that, The method includes: Acquire raw images captured by external cameras inside and outside the vehicle; Based on the original image, the distribution pattern of interference information in the original image is obtained; Based on the distribution pattern of the interference information, the target processing method is obtained; The interference information in the original image is processed based on the target processing method to obtain a processed image, and the processed image is displayed on the display screen of the vehicle.

2. The method according to claim 1, characterized in that, The target processing method obtained based on the distribution of the interference information includes: If the distribution of the interference information is element-level, then filtering will be determined as the target processing method. If the distribution of the interference information is regional, then the repair process will be determined as the target processing method. If the distribution of the interference information includes element-level distribution and region-level distribution, then filtering and repair processing will be determined as the target processing method.

3. The method according to claim 1, characterized in that, The step of processing the interference information in the original image based on the target processing method to obtain the processed image includes: If the target processing method is filtering, the interference information distributed at the element level in the original image is filtered to obtain the processed image; If the target processing method is repair processing, the interference information distributed at the region level in the original image is repaired to obtain the processed image; If the target processing method includes filtering and repair processing, the interference information distributed at the element level in the original image is filtered, and the interference information distributed at the region level in the original image is repaired to obtain the processed image.

4. The method according to claim 3, characterized in that, The restoration process for the region-level distributed interference information in the original image includes: Image content detection is performed on the image region where the regionally distributed interference information is located to obtain a detection result, which is used to indicate whether the image content of the image region is complete; Based on the detection results, the interference information distributed at the regional level in the original image is repaired.

5. The method according to claim 4, characterized in that, The restoration process, based on the detection results, for the interference information distributed at the region level in the original image, includes: If the detection result indicates that the image content is complete, the image display parameters of the image region are repaired, including brightness and / or contrast. If the detection result indicates that the image content is incomplete, the image content of the image region is repaired.

6. The method according to claim 5, characterized in that, The image display parameter repair process for the image region includes: Based on the current terrain features where the vehicle is located and the terrain feature database, the target repair parameters corresponding to the current terrain features are obtained. The terrain feature database includes repair parameters corresponding to different terrain features. Based on the target repair parameters, the image display parameters of the image region are repaired.

7. The method according to claim 5, characterized in that, Also includes: Retrieve pre-saved image frames; The image content restoration process for the image region includes: Based on the image content of the pre-saved image frame, the image content of the image region is repaired.

8. The method according to claim 7, characterized in that, Also includes: Based on the movement trajectory of the target object, the target camera in the vehicle that is obstructed is identified; The image frames captured by the target camera are saved.

9. The method according to any one of claims 1 to 8, characterized in that, The step of obtaining the distribution pattern of interference information in the original image based on the original image includes: Target detection and image segmentation are performed on the original image to obtain the distribution pattern of interference information in the original image.

10. An image processing apparatus, characterized in that, The device includes: Memory, used to store executable program code; A processor for calling and running the executable program code from the memory, causing the device to perform the method as described in any one of claims 1 to 9.

Images