Visual perception method, image preprocessing method, device, medium and vehicle thereof

Abstract

The present application relates to the technical field of automatic driving, and particularly provides a visual perception method, an image preprocessing method thereof, equipment, a medium and a vehicle, aiming to improve the image preprocessing efficiency. To this end, the method provided by the present application comprises: obtaining a to-be-perceived image of a visual perception task; performing reverse mapping on the to-be-perceived image by using a preset image reverse mapping relationship to obtain an image preprocessing result of the to-be-perceived image. The image reverse mapping relationship is a mapping relationship capable of realizing a preprocessing process represented by an image preprocessing link, which is obtained based on an image reverse mapping mechanism. The preprocessing process comprises sequentially performing a plurality of different image preprocessing operations on the image to obtain the image preprocessing result. The above method can greatly reduce the data read-write overhead, and does not need to repeatedly read and write data, thereby greatly improving the image preprocessing efficiency.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, specifically to a visual perception method and its image preprocessing method, device, medium, and vehicle. Background Technology

[0002] When controlling autonomous driving, cameras typically capture images of the vehicle's surroundings. These images are then used for visual perception to obtain information about various targets (such as obstacles and pedestrians) and their locations and trajectories. This information is then used to control the vehicle and ensure driving safety. Before visual perception, images typically undergo preprocessing operations such as color space conversion, cropping, scaling, and distortion correction to obtain a preprocessed image. However, current conventional image preprocessing methods mainly combine various preprocessing operations using existing open-source algorithms. The result of each preprocessing operation is stored separately as an intermediate result (written data) and then input (read data) to the next preprocessing operation. This repeated data reading and writing reduces the efficiency of image preprocessing. Autonomous driving has very high real-time requirements; if image preprocessing cannot be completed quickly and reliably, it will significantly reduce the safety of autonomous driving.

[0003] Accordingly, a new technical solution is needed in this field to solve the above problems. Summary of the Invention

[0004] To overcome the above-mentioned defects, the present invention is proposed to provide a visual perception method and its image preprocessing method, device, medium, and vehicle that solve or at least partially solve the technical problem of how to improve the efficiency of image preprocessing while ensuring the accuracy of image preprocessing.

[0005] In a first aspect, an image preprocessing method for visual perception is provided, the method comprising:

[0006] Acquire the image to be perceived for the visual perception task;

[0007] A preset image inverse mapping relationship is used to perform inverse mapping on the image to be perceived in order to obtain the image preprocessing result of the image to be perceived.

[0008] The image inverse mapping relationship is a mapping relationship obtained based on the image inverse mapping mechanism that can realize the preprocessing process represented by the image preprocessing link. The preprocessing process is to sequentially perform a variety of different image preprocessing operations on the image to obtain the image preprocessing result.

[0009] In the above-mentioned image preprocessing method for visual perception, the method further includes obtaining a preset image inverse mapping relationship through the following means:

[0010] Based on the type of visual perception task, obtain the image preprocessing link that matches the type from a set of preset image preprocessing links;

[0011] Obtain the image inverse mapping relationship corresponding to the matched image preprocessing link, with each image preprocessing link corresponding one-to-one with each preset image inverse mapping relationship.

[0012] In the above-described image preprocessing method for visual perception, obtaining the image preprocessing link matching the type includes:

[0013] Obtain the combination of operation types for image preprocessing operations corresponding to the aforementioned type;

[0014] Based on the operation type combination corresponding to the type, the corresponding image preprocessing link is obtained, and each operation type combination corresponds one-to-one with each image preprocessing link.

[0015] The operation type combination includes at least one operation type of image preprocessing operation, and the operation type included in the operation type combination is the same as the operation type of the image preprocessing operation performed in the preprocessing process represented by the corresponding image preprocessing link.

[0016] In the above-described image preprocessing method for visual perception, before obtaining the image preprocessing result of the image to be perceived, the method further includes:

[0017] Load the camera intrinsic parameters corresponding to the image to be perceived, so that the image to be perceived can be reverse mapped according to the camera intrinsic parameters and using a preset image reverse mapping relationship.

[0018] In the above-described image preprocessing method for visual perception, after obtaining the image preprocessing result, the method further includes:

[0019] The image preprocessing results are stored in vector form.

[0020] In a second aspect, a visual perception method is provided, the method comprising:

[0021] Acquire the image to be perceived for the visual perception task;

[0022] Using the image preprocessing method for visual perception provided in the first aspect above, the image preprocessing result of the image to be perceived is obtained;

[0023] Visual perception is performed based on the image preprocessing results.

[0024] In one technical solution of the above-mentioned visual perception method, the step of "performing visual perception based on the image preprocessing result" specifically includes:

[0025] Based on the type of visual perception task, a system-on-chip (SoC) corresponding to the type is obtained from a plurality of preset system-on-chips, and each SoC is used to execute the visual perception task of its corresponding type.

[0026] The image processing results are sent to the corresponding system-on-chip (SoC) for visual perception.

[0027] In a third aspect, a computer device is provided, the computer device including a processor and a storage device, the storage device being adapted to store a plurality of program codes, the program codes being adapted to be loaded and run by the processor to perform the method described in any of the above-described technical solutions for image preprocessing methods or visual perception methods.

[0028] In a fourth aspect, a computer-readable storage medium is provided, wherein a plurality of program codes are stored therein, the program codes being adapted to be loaded and run by a processor to perform the method described in any of the above-described image preprocessing method or visual perception method for visual perception.

[0029] In a fifth aspect, a vehicle is provided that includes the computer equipment described in the above-described computer equipment technical solution.

[0030] The above-described technical solutions of the present invention have at least one or more of the following beneficial effects:

[0031] In the technical solution of the image preprocessing method for visual perception provided by the present invention, the image to be perceived for the visual perception task can be acquired, and a preset image inverse mapping relationship can be used to perform inverse mapping on the image to be perceived in order to obtain the image preprocessing result of the image to be perceived. The image inverse mapping relationship is a mapping relationship obtained based on the image inverse mapping mechanism that can realize the preprocessing process represented by the image preprocessing link. The preprocessing process involves sequentially performing multiple different image preprocessing operations on the image to obtain the image preprocessing result.

[0032] In the above method, the results of each image preprocessing operation are not stored separately as intermediate results. The entire image preprocessing process involves only one data read (acquiring the image to be perceived) and only one data write (storing the image preprocessing result after it is acquired). Therefore, this method significantly reduces the overhead of data reading and writing, eliminating the need for repeated data reads and writes, thus greatly improving the efficiency of image preprocessing. Furthermore, the image preprocessing results obtained through inverse mapping based on the above image inverse mapping relationship have high accuracy, ensuring the accuracy of the image preprocessing results.

[0033] In the technical solution of the visual perception method of the present invention, an image to be perceived for the visual perception task can be acquired, and the aforementioned image preprocessing method for visual perception can be used to obtain the image preprocessing result of the image to be perceived. Visual perception is then performed based on the image preprocessing result. This method can quickly and accurately complete image preprocessing, thereby significantly improving the accuracy and efficiency of visual perception. Attached Figure Description

[0034] The disclosure of this invention will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. Wherein:

[0035] Figure 1 This is a schematic flowchart of the main steps of an image preprocessing method for visual perception according to an embodiment of the present invention.

[0036] Figure 2 This is a schematic flowchart of the main steps of a method for obtaining an image inverse mapping relationship according to an embodiment of the present invention;

[0037] Figure 3 This is a schematic diagram of multiple image preprocessing links according to an embodiment of the present invention;

[0038] Figure 4 This is a schematic flowchart of the main steps of a visual perception method according to an embodiment of the present invention;

[0039] Figure 5 This is a schematic diagram of the main structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0040] Some embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0041] In the description of this invention, "processor" can include hardware, software, or a combination of both. A processor can be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and / or signal processing capabilities. The processor can be implemented in software, in hardware, or a combination of both. Computer-readable storage media includes any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc.

[0042] The following describes an embodiment of the image preprocessing method provided by the present invention.

[0043] See appendix Figure 1 , Figure 1 This is a schematic flowchart illustrating the main steps of an image preprocessing method for visual perception according to an embodiment of the present invention. Figure 1 As shown, the image preprocessing method in this embodiment of the invention mainly includes the following steps S101 to S102.

[0044] Step S101: Obtain the image to be perceived for the visual perception task.

[0045] The image to be perceived can be acquired through a camera on the vehicle. It should be noted that all vehicle-related operations, such as acquiring image frames through a vehicle's camera and performing autonomous driving control, mentioned in this invention, are executed only after full authorization from the user or other parties. In other words, the vehicle in this invention is an authorized vehicle. In some embodiments, the vehicle's infotainment system or a backend server can detect whether authorization information has been received. If authorization information is received, it indicates that the current vehicle is authorized; otherwise, the current vehicle is unauthorized. The authorization information can be sent through terminal devices including, but not limited to, mobile phones, tablets, and smartwatches.

[0046] Step S102: Using a preset image inverse mapping relationship, perform inverse mapping on the image to be perceived to obtain the image preprocessing result of the image to be perceived.

[0047] Image inverse mapping is a mapping relationship obtained based on the image inverse mapping mechanism that can realize the preprocessing process represented by the image preprocessing link. The preprocessing process involves sequentially performing various image preprocessing operations on the image to obtain the image preprocessing result.

[0048] Image preprocessing operations include, but are not limited to, color space conversion, image distortion correction, image cropping, image scaling, and image numerical type conversion. Those skilled in the art can flexibly define the operation content of various image preprocessing operations according to actual needs; this embodiment of the invention does not impose specific limitations on this.

[0049] Image inverse mapping is a common inverse mapping mechanism applied to images in the field of vision technology. The following is a simple explanation of the image inverse mapping mechanism using the single image preprocessing operation of obtaining grayscale values ​​as an example. Assume we have an input image and need to obtain the output image. For each position (x, y) in the output image, we can use [v, w, 1] = [x, y, 1] * T. -1 Calculate the position (v, w) of the given location (x, y) in the input image, then calculate the grayscale value at position (v, w) in the input image using interpolation. Use this interpolated grayscale value as the grayscale value at position (x, y) in the output image. Let T represent the transformation matrix between the input and output images. -1 This represents the inverse matrix of the transformation matrix. The interpolation methods include, but are not limited to, bilinear interpolation.

[0050] In this embodiment of the invention, the preprocessing process involves sequentially performing multiple different image preprocessing operations on an image. Therefore, the image inverse mapping relationship obtained based on the image inverse mapping mechanism is not a single image preprocessing operation inverse mapping relationship, but rather a fusion mapping relationship that simultaneously integrates the inverse mapping relationships between different image preprocessing operations and the fusion mapping relationship based on the execution order of different image preprocessing operations. For example, if the image preprocessing operations executed sequentially in the image preprocessing chain are color space conversion, image scaling, and image cropping, then the image inverse mapping relationship can be represented as f3(f2(f1(image)), where f1 represents the inverse mapping relationship for color space conversion of the input image image, f2 represents the inverse mapping relationship for image scaling of the color space conversion result f1(image), and f3 represents the inverse mapping relationship for image cropping of the scaling result f2(f1(image). The result obtained after f3(f2(f1(image))) is the image preprocessing result.

[0051] In some implementations, different images to be perceived may be acquired by different cameras, and different cameras may have different camera intrinsic parameters. Therefore, before obtaining the image preprocessing result of the image to be perceived, the camera intrinsic parameters corresponding to the image to be perceived can be loaded first. When executing step S102, the image to be perceived is reverse-mapped according to the loaded camera intrinsic parameters and a preset image inverse mapping relationship to obtain the image preprocessing result of the image to be perceived. Camera intrinsic parameters are parameters related to the characteristics of the camera itself, such as the camera's focal length and pixel size. Different camera intrinsic parameters may be used when performing different image preprocessing operations. Therefore, before using the image inverse mapping relationship to reverse-map the image to be perceived, the relevant camera intrinsic parameters must also be obtained first, and then the inverse mapping is performed in combination with the camera intrinsic parameters to realize the image preprocessing operation.

[0052] Furthermore, in some embodiments, after obtaining the image preprocessing results, to facilitate rapid visual perception by the computer device performing the visual perception task, the image preprocessing results can be stored in vector form. This eliminates the need to convert the preprocessed image into vector form during visual perception, thus omitting this step and improving the efficiency of visual perception. It should be noted that those skilled in the art can flexibly define the specific representation of the vector according to actual needs, and this embodiment of the invention does not impose specific limitations on this. For example, each element in the vector can correspond one-to-one with each pixel, and the value of the element can include various information such as the pixel's coordinates and grayscale.

[0053] In the method described in steps S101 to S102 above, the result of each image preprocessing operation is not stored separately as an intermediate result. The entire image preprocessing process involves only one data read (acquiring the image to be perceived) and only one data write (storing the image preprocessing result after it is acquired). Therefore, the above method does not require repeated data reading and writing, thus greatly improving the efficiency of image preprocessing. Furthermore, the image preprocessing result obtained by reverse mapping based on the above image reverse mapping relationship has high accuracy, ensuring the accuracy of the image preprocessing result.

[0054] The method for obtaining a preset image reverse mapping relationship in an embodiment of the present invention will be described below. (See appendix.) Figure 2 In this embodiment of the invention, the image reverse mapping relationship used in step S102 can be obtained through the following steps S201 to S202.

[0055] Step S201: Based on the type of visual perception task, obtain the image preprocessing link that matches the type from a set of preset image preprocessing links.

[0056] Taking obstacle perception in visual perception tasks as an example, the types of visual perception tasks can include perceiving the position, trajectory, and speed of obstacles. Those skilled in the art can flexibly classify visual perception tasks according to actual needs. This embodiment of the invention does not specifically limit the classification method of visual perception tasks, as long as the type of the visual perception task can be obtained.

[0057] In this embodiment of the invention, multiple image preprocessing links can be set up, and a matching relationship between different visual perception task types and different image preprocessing links can be established in advance. When it is necessary to obtain the image preprocessing link, the matching relationship can be retrieved, and the type of visual perception task can be matched to obtain the image preprocessing link that matches the type.

[0058] Step S202: Obtain the image inverse mapping relationship corresponding to the matched image preprocessing link. In this embodiment of the invention, each image preprocessing link represents a different preprocessing process, and the image inverse mapping relationship corresponding to each image preprocessing link will also be different. Therefore, each image preprocessing link can be pre-set to correspond one-to-one with each preset image inverse mapping relationship. When it is necessary to obtain the image inverse mapping relationship, it is only necessary to call this correspondence, and the corresponding image inverse mapping relationship can be obtained using this correspondence.

[0059] Based on the methods described in steps S201 to S202 above, multiple different image preprocessing links can be set up to meet various image preprocessing requirements. Simultaneously, the image inverse mapping relationship to be used can be quickly and accurately obtained from the image inverse mapping relationships corresponding to these image preprocessing links according to the type of visual perception task, further improving the efficiency and accuracy of image preprocessing.

[0060] The following provides a further explanation of step S201.

[0061] In this embodiment of the invention, the image preprocessing link matching the type of visual perception task can be obtained through the following steps S2011 to S2012.

[0062] Step S2011: Obtain the combination of operation types for image preprocessing operations corresponding to the type of visual perception task.

[0063] In this embodiment of the invention, multiple operation type combinations can be set, and a matching relationship between different visual perception task types and different operation type combinations can be established in advance. When it is necessary to obtain the operation type combination, the matching relationship can be retrieved, and the type of visual perception task can be matched to obtain the operation type combination that matches the type.

[0064] Step S2012: Obtain the corresponding image preprocessing link based on the combination of operation types corresponding to the type of visual perception task.

[0065] In this embodiment of the invention, each operation type combination corresponds one-to-one with each image preprocessing link. This correspondence can be pre-set, and when it is necessary to obtain an image preprocessing link, this correspondence only needs to be invoked to obtain the corresponding image preprocessing link. The operation type combination includes at least one operation type of image preprocessing operation, and the operation type included in the operation type combination is the same as the operation type of the image preprocessing operation performed in the preprocessing process represented by the corresponding image preprocessing link.

[0066] For example, suppose there are four combinations of operation types: the first includes image distortion correction, image cropping, image scaling, and image data type conversion; the second includes color space conversion, image distortion correction, image cropping, image scaling, and image data type conversion; the third includes color space conversion, image distortion correction, image scaling, and image data type conversion; and the fourth includes image data type conversion. Figure 3 As shown, the combination of the above four operation types will form four image preprocessing links, namely... Figure 3 Link 1, Link 2, Link 3, and Link 4 in the diagram, with the first to fourth operation type combinations corresponding one-to-one with Link 1 to Link 4 respectively.

[0067] Based on the methods described in steps S2011 to S2012 above, the image preprocessing link that needs to be used can be obtained quickly and accurately according to the type of visual perception task, thereby further improving the efficiency and accuracy of image preprocessing.

[0068] The following describes embodiments of the visual perception method provided by the present invention.

[0069] See appendix Figure 4 , Figure 4 This is a schematic flowchart illustrating the main steps of a visual perception method according to an embodiment of the present invention. Figure 4 As shown, the visual perception method in this embodiment of the invention mainly includes the following steps S301 to S303.

[0070] Step S301: Acquire the image to be perceived for the visual perception task.

[0071] Step S302: Use an image preprocessing method for visual perception to obtain the image preprocessing result of the image to be perceived. The image preprocessing method in this step is the image preprocessing method described in the foregoing method embodiments.

[0072] Step S303: Perform visual perception based on the image preprocessing results.

[0073] Based on the methods described in steps S301 to S303 above, image preprocessing can be completed quickly and accurately, thereby significantly improving the accuracy and efficiency of visual perception.

[0074] The following provides a further explanation of step S303.

[0075] In this embodiment of the invention, visual perception can be performed based on the image preprocessing results through the following steps S3031 to S3032.

[0076] Step S3031: Based on the type of visual perception task, obtain the corresponding system-on-chip (SoC) from a pre-set pool of multiple SoCs. Each SoC is used to execute its respective type of visual perception task. Step S3032: Send the image processing result to the SoC obtained in step S3031 for visual perception.

[0077] like Figure 3 As shown, assuming the visual perception task is to perceive the location of obstacles, the corresponding system-on-a-chip (SoC) is SoC 1. If among multiple image preprocessing links, the image preprocessing link matching this type is... Figure 3 Link 1 in the process. After obtaining the image preprocessing result through Link 1, the image preprocessing result is sent to SoC 1. If, among multiple image preprocessing links matching this type, there is also Link 2 in addition to Link 1, then the image preprocessing result obtained from Link 2 should also be sent to SoC 1.

[0078] Assume the visual perception task includes perceiving the position and trajectory of an obstacle, with corresponding system-on-chips (SoCs) 1 and SoC 2, respectively. If link 1 in the multiple image preprocessing links matches the perceived obstacle position, the image preprocessing result obtained from link 1 is sent to SoC 1; if link 2 in the multiple image preprocessing links matches the perceived obstacle trajectory, the image preprocessing result obtained from link 2 is sent to SoC 2. This achieves... Figure 3 The cross-SoC transfer shown.

[0079] Using the above method, even if there are multiple SoCs and each SoC has different requirements for image preprocessing, the image preprocessing results that meet the requirements of each SoC can be quickly sent to each SoC, thereby further improving the efficiency of visual perception.

[0080] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effects of the present invention, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders. These adjusted solutions are equivalent to the technical solutions described in the present invention and therefore will also fall within the protection scope of the present invention.

[0081] Those skilled in the art will understand that all or part of the processes in the method of the above embodiment of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or device capable of carrying the computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the content included in the computer-readable storage medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0082] Furthermore, the present invention also provides a computer device.

[0083] See appendix Figure 5 , Figure 5 This is a schematic diagram of the main structure of a computer device according to an embodiment of the present invention. Figure 5 As shown, the computer device in this embodiment of the invention mainly includes a storage device and a processor. The storage device can be configured to store a program for executing the image preprocessing method or visual perception method for visual perception described in the above method embodiments. The processor can be configured to execute the program in the storage device, which includes, but is not limited to, the program for executing the image preprocessing method or visual perception method for visual perception described in the above method embodiments. For ease of explanation, only the parts related to the embodiments of the present invention are shown. For specific technical details not disclosed, please refer to the method section of the embodiments of the present invention.

[0084] In embodiments of the present invention, the computer device may be a control device comprising various electronic devices. In some possible implementations, the computer device may include multiple storage devices and multiple processors. The program executing the image preprocessing method or visual perception method for visual perception described in the above method embodiments may be divided into multiple subroutines. Each subroutine may be loaded and run by a processor to execute different steps of the image preprocessing method or visual perception method for visual perception described in the above method embodiments. Specifically, each subroutine may be stored in different storage devices, and each processor may be configured to execute programs in one or more storage devices to jointly implement the image preprocessing method or visual perception method for visual perception described in the above method embodiments. That is, each processor executes different steps of the image preprocessing method or visual perception method for visual perception described in the above method embodiments to jointly implement the image preprocessing method or visual perception method for visual perception described in the above method embodiments.

[0085] The aforementioned multiple processors can be processors deployed on the same device. For example, the aforementioned computer device can be a high-performance device composed of multiple processors, and the aforementioned multiple processors can be processors configured on that high-performance device. Alternatively, the aforementioned multiple processors can also be processors deployed on different devices. For example, the aforementioned computer device can be a server cluster, and the aforementioned multiple processors can be processors on different servers within the server cluster.

[0086] Furthermore, the present invention also provides a computer-readable storage medium.

[0087] In one embodiment of a computer-readable storage medium according to the present invention, the computer-readable storage medium may be configured to store a program for performing the image preprocessing method or visual perception method for visual perception described in the above-described method embodiments. This program may be loaded and run by a processor to implement the image preprocessing method or visual perception method for visual perception described above. For ease of explanation, only the parts related to the embodiments of the present invention are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of the present invention. The computer-readable storage medium may be a storage device comprising various electronic devices. Optionally, in the embodiments of the present invention, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0088] Furthermore, the present invention also provides a vehicle.

[0089] In one embodiment of a vehicle according to the present invention, the vehicle may include the computer equipment described in the above-described computer equipment embodiments. In this embodiment, the vehicle may be an autonomous vehicle, an unmanned vehicle, or the like. Furthermore, according to the type of power source, the vehicle in this embodiment may be a gasoline vehicle, an electric vehicle, a hybrid vehicle that combines electric and gasoline power, or a vehicle using other new energy sources, etc.

[0090] The technical solution of the present invention has been described above with reference to one embodiment shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions resulting from such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for pre-processing an image for visual perception, characterized in that, The method includes: Acquire the image to be perceived for the visual perception task; A preset image inverse mapping relationship is used to perform inverse mapping on the image to be perceived in order to obtain the image preprocessing result of the image to be perceived. The image inverse mapping relationship is a mapping relationship obtained based on the image inverse mapping mechanism that can realize the preprocessing process represented by the image preprocessing link. The preprocessing process is to sequentially perform a variety of different image preprocessing operations on the image to obtain the image preprocessing result. The method further includes obtaining a preset image inverse mapping relationship through the following means: Based on the type of visual perception task, obtain the image preprocessing link that matches the type from a set of preset image preprocessing links; Obtain the image inverse mapping relationship corresponding to the matched image preprocessing link, with each image preprocessing link corresponding one-to-one with each preset image inverse mapping relationship.

2. The image pre-processing method of claim 1, wherein, Obtaining the image preprocessing link matching the type includes: Obtain the combination of operation types for image preprocessing operations corresponding to the aforementioned type; Based on the operation type combination corresponding to the type, the corresponding image preprocessing link is obtained, and each operation type combination corresponds one-to-one with each image preprocessing link. The operation type combination includes at least one operation type of image preprocessing operation, and the operation type included in the operation type combination is the same as the operation type of the image preprocessing operation performed in the preprocessing process represented by the corresponding image preprocessing link.

3. The image pre-processing method of claim 1, wherein, Before obtaining the image preprocessing result of the image to be perceived, the method further includes: Load the camera intrinsic parameters corresponding to the image to be perceived, so that the image to be perceived can be reverse mapped according to the camera intrinsic parameters and using a preset image reverse mapping relationship.

4. The image pre-processing method of claim 1, wherein, After obtaining the image preprocessing results, the method further includes: The image preprocessing results are stored in vector form.

5. A method of visual perception, characterized by, The method includes: Acquire the image to be perceived for the visual perception task; The image preprocessing method for visual perception according to any one of claims 1 to 4 is used to obtain the image preprocessing result of the image to be perceived. Visual perception is performed based on the image preprocessing results.

6. The visual perception method of claim 5, wherein, The step of "performing visual perception based on the image preprocessing results" specifically includes: Based on the type of visual perception task, a system-on-a-chip corresponding to the type is obtained from a plurality of preset system-on-a-chips, and each system-on-a-chip is used to execute the visual perception task of its corresponding type. The image processing results are sent to the corresponding system-on-a-chip for visual perception.

7. A computer device comprising a processor and a storage device, said storage device being adapted to store a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by the processor to perform the image preprocessing method for visual perception as described in any one of claims 1 to 4, or to perform the visual perception method as described in any one of claims 5 to 6.

8. A computer readable storage medium having stored therein a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by a processor to perform the image preprocessing method for visual perception as described in any one of claims 1 to 4, or to perform the visual perception method as described in any one of claims 5 to 6.

9. A vehicle characterized by comprising: The vehicle includes the computer equipment as described in claim 7.

Images