A data processing method, system and apparatus

Abstract

The application provides a data processing method, system and device, relates to the field of AI, and the data processing method can be executed by a server or a device with a data processing function. In execution, reference data is acquired first. The reference data includes RGB image data and device parameters of an image device. Then, AutoML is used to determine a plurality of conversion parameters required for converting the RGB image data into RAW data. Finally, the RGB image data is processed into RAW data according to the plurality of conversion parameters. The RAW data matches the device parameters of the image device. Since the application introduces AutoML to determine the plurality of conversion parameters required for converting the RGB image data into RAW data, instead of determining the conversion parameters based on artificial experience, the efficiency of data processing is improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence (AI) technology, and in particular to a data processing method, system and apparatus. Background Technology

[0002] Computational photography is a technology that integrates computer software methods (computer vision, digital signal processing, graphics, etc.) with photographic equipment and related applications. By organically combining hardware design with software computing power, computational photography greatly simplifies the photographic process, enhances the photographic experience, and allows more people to enjoy the pleasures of photography.

[0003] To achieve good image quality on terminal devices (smartphones, tablets, etc.), a RAW-RGB dataset, composed of RAW data (all grayscale data of the image recorded by the image sensor) generated during the shooting process of the terminal device and RGB data generated by the SLR camera, can be used for neural network learning to improve the shooting capabilities of the terminal device. However, acquiring RAW-RGB datasets is time-consuming, and the acquired RAW-RGB dataset differs significantly from the real data.

[0004] Considering the problems that arise during data acquisition, related technologies propose constructing RAW-RGB datasets by degrading RGB data to RAW data. However, this method requires manual calculation of the values ​​of multiple parameters needed to degrade RGB data to RAW data, which is time-consuming and labor-intensive. Summary of the Invention

[0005] Based on this, this application provides a data processing method, system, and apparatus to improve data processing efficiency.

[0006] Firstly, this application provides a data processing method that can be executed by a server or by a device with data processing capabilities, without specific limitations. When executed, this method first obtains reference data, which includes RGB image data and the device parameters of the image device. Then, it uses automated machine learning (AutoML) to determine multiple conversion parameters required to convert the RGB image data into RAW data. Finally, it processes the RGB image data into RAW data based on these multiple conversion parameters, and the RAW data is matched with the device parameters of the image device.

[0007] The aforementioned image device can be understood as a device with image processing capabilities. It can be a terminal device such as a mobile phone, tablet computer, or mobile robot, or other devices; this application does not specifically limit it. This application utilizes AutoML to determine multiple conversion parameters required for converting RGB image data into RAW data, and converts the RGB image data into RAW data based on these conversion parameters. RAW data obtained in this way can improve data processing efficiency, and the RAW data matches the device parameters of the image device, making it more suitable for the device's requirements.

[0008] In one possible implementation, a search space corresponding to the image device can be determined based on device parameters; then, multiple transformation parameters can be determined from the search space.

[0009] It should be noted that different image devices have different device parameters. This application can adapt a search space that meets the needs of the image device according to the device parameters of different image devices, and determine the conversion parameters that meet the needs of the image device within the determined search space. The conversion parameters determined in this way can improve the efficiency of obtaining conversion parameters, and the conversion parameters that meet the needs of the image device are conducive to better converting RGB image data into RAW data.

[0010] In one possible implementation, image pairs of RGB image data and RAW data can be constructed; then, the image pairs are input into a task processing unit for training to determine feedback signals; the task processing unit is used to process video or image data; the feedback signals are used to indicate the construction quality of the image pairs; finally, multiple conversion parameters required for converting RGB image data into RAW data are updated based on the feedback signals.

[0011] It should be noted that the above-mentioned construction quality reflects the degree of matching between the RAW data and the device parameters of the image device when converting RGB image data to RAW data. This application adjusts the multiple conversion parameters required for converting RGB image data to RAW data based on feedback signals, thereby improving the accuracy of the conversion parameters and making the RAW data more suitable for the device parameter requirements of the image device.

[0012] In one possible implementation, the search space corresponding to the image device includes multiple image processing modules; wherein, the image processing modules include one or more of the following: a noise addition module, a mosaic addition module, and a brightness adjustment module.

[0013] It should be noted that the image processing module in this application is not limited to a noise addition module, a mosaic addition module, and a brightness adjustment module, but may also include a level adjustment module, a white balance adjustment module, etc. Furthermore, any image processing module required for converting other RGB image data to RAW data is applicable to this application, and this application does not specifically limit the number and type of image processing modules included.

[0014] In one possible implementation, the conversion parameters include one or more of the following: noise addition parameters, mosaic parameters, brightness adjustment parameters, gamma parameters, level adjustment parameters, and white balance adjustment parameters.

[0015] It should be noted that the conversion parameters in this application are not limited to noise addition parameters, mosaic parameters, brightness adjustment parameters, gamma parameters, level adjustment parameters, and white balance adjustment parameters, and may also include bad pixel correction parameters, etc. Furthermore, the conversion parameters required for converting other RGB image data to RAW data are also applicable to this application, and this application does not specifically limit the number and type of conversion parameters included.

[0016] Secondly, this application provides a data processing system, including: an image degradation unit and a strategy unit; the image degradation unit is used to convert RGB image data into RAW data that matches the device parameters of an image device according to multiple conversion parameters output by the strategy unit; the strategy unit is used to determine multiple conversion parameters required for converting the RGB image data into RAW data using automated machine learning AutoML.

[0017] It should be noted that, in practical applications, the aforementioned image degradation unit and strategy unit can be understood as an image degradation device and strategy device, or as an image degradation model and strategy model; this application does not specifically limit these interpretations. In this application, the degradation unit can utilize AutoML to determine multiple conversion parameters required for converting RGB image data into RAW data, and the image degradation unit can convert the RGB image data into RAW data based on these conversion parameters. The RAW data obtained through this data processing system can improve data processing efficiency, and the RAW data matches the device parameters of the image device, making it more suitable for the device requirements of the image device.

[0018] In one possible implementation, the image degradation unit is also used to output an image pair of the RGB image data and the RAW data.

[0019] In one possible implementation, the system further includes a task processing unit; the task processing unit is used to train the image pairs of the RGB image data and the RAW data output by the image degradation unit, determine a feedback signal, and input the feedback signal to the strategy unit; the feedback signal is used to indicate the construction quality of the image pairs.

[0020] In one possible implementation, the strategy unit is further configured to receive the feedback signal output by the task processing unit, and adjust the network parameters of the strategy unit according to the feedback signal; and update the plurality of transformation parameters based on the adjusted network parameters.

[0021] It should be noted that the feedback signal in this application is used to indicate the construction quality of the image pairs of generated RGB image data and RAW data. Adjusting the network parameters based on the feedback signal and updating multiple conversion parameters required for converting RGB image data into RAW data based on the adjusted network parameters can make the RAW data more suitable for the device parameter requirements of the image device.

[0022] In one possible implementation, the strategy unit is further configured to determine a search space corresponding to the image device based on the device parameters; and to determine the plurality of transformation parameters from the search space.

[0023] It should be noted that different image devices have different device parameters. This application can adapt a search space that meets the needs of the image device according to the device parameters of different image devices, and determine the conversion parameters that meet the needs of the image device within the determined search space. The conversion parameters determined in this way can improve the efficiency of obtaining conversion parameters, and the conversion parameters that meet the needs of the image device are conducive to better converting RGB image data into RAW data.

[0024] In one possible implementation, the image degradation unit includes multiple image processing modules; wherein the image processing modules include one or more of the following: a noise addition module, a mosaic addition module, and a brightness adjustment module.

[0025] It should be noted that the image processing module in this application is not limited to a noise addition module, a mosaic addition module, and a brightness adjustment module, but may also include a level adjustment module, a white balance adjustment module, etc. Furthermore, any image processing module required for converting other RGB image data to RAW data is applicable to this application, and this application does not specifically limit the number and type of image processing modules included.

[0026] In one possible implementation, the conversion parameters include one or more of the following: noise addition parameters, mosaic addition parameters, brightness adjustment parameters, gamma parameters, level adjustment parameters, and white balance adjustment parameters.

[0027] It should be noted that the conversion parameters in this application are not limited to noise addition parameters, mosaic parameters, brightness adjustment parameters, gamma parameters, level adjustment parameters, and white balance adjustment parameters, and may also include bad pixel correction parameters, etc. Furthermore, the conversion parameters required for converting other RGB image data to RAW data are also applicable to this application, and this application does not specifically limit the number and type of conversion parameters included.

[0028] Thirdly, this application provides a data processing apparatus, including a processor and a memory, wherein the memory stores a computer program; the processor is configured to execute the computer program stored in the memory so that the solution described in any implementation of the first aspect is executed.

[0029] Fourthly, this application provides a computer-readable storage medium storing computer-readable instructions, which, when read and executed by a computer, cause the computer to perform the solution described in any implementation of the first aspect above.

[0030] Fifthly, this application provides a computer program product that, when read and executed by a computer, causes the computer to perform the scheme described in any of the implementations of the first aspect above.

[0031] For the technical effects that can be achieved by the second to fifth aspects mentioned above, please refer to the description of the technical effects that can be achieved by the corresponding possible design schemes in the first aspect mentioned above. This application will not repeat them here. Attached Figure Description

[0032] Figure 1 This diagram illustrates the structure of the data processing network provided in an embodiment of this application.

[0033] Figure 2 This diagram illustrates the structure of the image degradation unit provided in an embodiment of this application.

[0034] Figure 3 This illustration shows a schematic diagram of the structure of another data processing network provided in an embodiment of this application;

[0035] Figure 4 A flowchart illustrating the data processing method provided in an embodiment of this application is shown;

[0036] Figure 5 A schematic diagram of the structure of the data processing apparatus provided in an embodiment of this application is shown. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0038] As described in the background section, while related technologies propose constructing RAW-RGB datasets by degrading RGB data to RAW data, this method requires manual calculation of the values ​​of multiple parameters needed for the RGB data to be degraded to RAW data. For example, the RGB data to RAW data degrades to RAW data through image processing steps such as adding noise and adding mosaic. Assuming there are 100 parameters related to adding noise and 200 parameters related to adding mosaic, the RGB data to RAW data degrades to a maximum of 20,000 times (100*200) to determine the parameter values ​​required. This method is time-consuming and labor-intensive, and the RAW data generated by this method may not be suitable for the device requirements of different image devices. Therefore, there is an urgent need for a data processing method to solve the above problems.

[0039] It should be noted that in this application, "and / or" describes 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, or B existing alone, where A and B can be singular or plural. Furthermore, unless otherwise stated, the ordinal numbers such as "first" and "second" mentioned in the embodiments of this application are used to distinguish multiple objects and are not used to limit the order, sequence, priority, or importance of multiple objects.

[0040] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in still other embodiments" appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments, but rather a specific feature, structure, or characteristic described in connection with that embodiment."

[0041] To better illustrate the solution of this application, a brief introduction to some technical terms used in this application is provided below:

[0042] AutoML is a process that applies machine learning to automate data processing for real-world problems. In practical applications, it can provide users with different parameters based on the specific task. For example, if the task is to find the prime numbers among 1000 natural numbers, and the 1000 natural numbers and the task of finding prime numbers are used as input data for AutoML, then all the prime numbers among the 1000 natural numbers can be output. If the task is to classify 10,000 images into portraits, natural landscapes, and cultural landscapes, and the 10,000 images are used as input data for AutoML, then the images can be output according to the different categories.

[0043] RGB Image: An RGB image consists of a three-dimensional array of format M×N×3, where "3" can be understood as three M×N two-dimensional images (grayscale images). These three images represent the R, G, and B components, respectively, with each component's pixel value ranging from [0, 255]. R represents Red, G represents Green, and B represents Blue. When defining color, the values ​​of the R, G, and B components range from 0 to 255, where 0 indicates no stimulus and 255 indicates the maximum stimulus. Specifically, when R, G, and B are all 255, white light is synthesized; when R, G, and B are all 0, black light is formed.

[0044] RAW data can be understood as a file that records the raw information from the image device's sensor, as well as metadata generated by the image device (such as ISO settings, shutter speed, aperture value, white balance, etc.).

[0045] like Figure 1 The schematic diagram of the data processing system provided in this application includes an image degradation unit and a strategy unit. The data processing system can be integrated into one device or multiple devices. If integrated into one device, the image degradation unit can be an image degradation model, and the strategy unit can be a model for outputting a strategy. If integrated into multiple devices, the image degradation unit can be a processing device for processing image degradation data, and the strategy unit can be a processing device for outputting an image processing strategy. This application does not specifically limit whether the unit is a model or a device.

[0046] It should be noted that the strategy unit can use AutoML to determine multiple conversion parameters required to convert RGB image data into RAW data; the image degradation unit can receive RGB image data and the device parameters of the image device, and convert the RGB image data into RAW data that matches the device parameters of the image device according to the multiple conversion parameters output by the strategy unit. Here, the image device can be understood as a device with image processing capabilities, which can be a terminal device such as a mobile phone, tablet computer, or mobile robot, or other devices; this application does not specifically limit this. The RGB image data can be an image directly downloaded from the network by the data processing system, an image captured by a DSLR camera and then imported into the data processing system, or an image downloaded from the network by the image device and transmitted to the data processing system; this application does not specifically limit the data source of the RGB image data.

[0047] This application utilizes AutoML to determine multiple conversion parameters required for converting RGB image data into RAW data, and then converts the RGB image data into RAW data based on these parameters. The RAW data obtained in this way improves data processing efficiency, and the RAW data determined in this application matches the device parameters of the image device, making it more suitable for the device requirements.

[0048] Furthermore, the strategy unit in this application can also determine a search space corresponding to the image device based on the device parameters; and determine multiple transformation parameters from the search space. Based on the transformation parameters determined by the strategy unit, the requirements of the image device's device parameters can be adaptively adapted. For example, if the ISO is parameter 1 and the aperture value is parameter 2 in the device parameters of image device 1, the server can determine a search space 1 adapted to image device 1 based on parameter 1 and parameter 2, and determine multiple transformation parameters adapted to the requirements of image device 1 within the search space 1; if the ISO is parameter 2, the aperture value is parameter 4, and the shutter speed is parameter 3 in the device parameters of image device 2, the server can determine a search space 2 adapted to image device 2 based on parameter 1, parameter 2, and parameter 3, and determine multiple transformation parameters adapted to the requirements of image device 2 within the search space 2.

[0049] Furthermore, the image degradation unit may include multiple image processing modules; wherein, the image processing modules may include one or more of the following: a noise addition module, a mosaic addition module, and a brightness adjustment module. The conversion parameters may include one or more of the following: noise addition parameters, mosaic addition parameters, brightness adjustment parameters, gamma parameters, level adjustment parameters, and white balance adjustment parameters.

[0050] It should be noted that since the image degradation unit of this application converts RGB image data and RAW data, that is, converts a high-definition image into a low-definition image, it is necessary to perform image processing procedures such as adding noise, adding mosaic, and adjusting brightness. However, in actual applications, it may not be limited to the above-mentioned image processing modules and may include other image processing modules. All image processing modules that may be involved in the process of converting RGB image data and RAW data are applicable to this application, and will not be illustrated one by one here.

[0051] Additionally, conversion parameters matching the image processing module may include noise addition parameters, mosaic parameters, brightness adjustment (e.g., increasing brightness, decreasing brightness) parameters, gamma parameters, level adjustment (e.g., adding black level, adding white level) parameters, and white balance adjustment parameters. Noise may include Gaussian white noise, additive noise, and multiplicative noise; gamma parameters can be used for grayscale adjustment and transparency adjustment; white balance can be used to eliminate the influence of light sources on the image, such as increasing the influence of light sources on the image by adjusting white balance adjustment parameters. All conversion parameters that may be involved in the process of converting RGB image data to RAW data are applicable to this application, and are not illustrated individually here.

[0052] It should be noted that the search space is related to the image processing modules in the image degradation unit, and the parameters corresponding to each image processing module in the image degradation unit (these parameters are matched with the device parameters of the image device). For example... Figure 2As shown, the conventional image processing flow involves removing noise from RAW data using a denoising module and removing mosaic effects from RAW data using a de-mosaic module, converting the RAW data into RGB image data. The image degradation processing in this application involves adding noise to the RGB image data using a noise-adding module and adding mosaic effects to the RGB image data using a mosaic-adding module, converting the RGB image data back into RAW data. This involves image processing such as adding noise and mosaic effects to the RGB image data (i.e., the opposite of converting RAW data into RGB image data). In practical applications, the image degradation unit may also include other image processing modules, which are not illustrated here. The search space mentioned in this application is related to each image processing module in the image degradation unit. For example, if the image degradation unit for the adapted image device 1 includes three image processing modules, such as a noise-adding module, a mosaic-adding module, and a brightness adjustment module, then the search space for the adapted image device 1 can be determined by the strategy unit. This search space includes parameters corresponding to the noise-adding module, the mosaic-adding module, and the brightness adjustment module. Furthermore, the conversion parameters corresponding to the noise addition module, mosaic addition module, and brightness adjustment module may include multiple parameters. The value range of the conversion parameters corresponding to the noise addition module, mosaic addition module, and brightness adjustment module can be determined according to the device parameters of the image device. For example, if the aperture value of image device 1 is A, the strategy unit knows that the value range of the conversion parameter corresponding to the brightness adjustment module is 50-500, not 0 to infinity, based on the aperture value A. Therefore, the parameter search range of the strategy unit can be narrowed. The relationship between the parameters of other image processing modules and the image device parameters will not be described in detail here.

[0053] As an example, a data processing system may also include task processing units, such as... Figure 3As shown, the task processing unit can be trained based on image pairs of RGB image data and RAW data output by the image degradation unit to determine a feedback signal. This feedback signal is then input to the policy unit, allowing the policy unit to adjust its network parameters based on the feedback signal and update multiple transformation parameters accordingly. It should be noted that the policy unit can be understood as a neural network model. Building this model requires a large number of network parameters related to its structure. Adjusting these network parameters can adjust the output policy of the policy unit, thereby adjusting the multiple transformation parameters output by the policy unit. The feedback signal indicates the construction quality of the generated image pairs. This feedback signal can be indicated by the loss value of the test dataset or training dataset. The construction quality reflects the degree of matching between the RAW data and the device parameters of the image device when converting RGB image data to RAW data. This task processing unit can be used to process video or image data, such as converting low-resolution data to ultra-high-definition video, low-resolution images to high-resolution images, or low-brightness images to high-brightness images, etc. This application does not specifically limit its capabilities.

[0054] For example, image pairs of RGB and RAW image data are input into the task processing unit for training. The feedback signal is then input into the policy unit. The policy unit adaptively adjusts the transformation parameters to be output based on the feedback signal and inputs these transformation parameters into the image degradation unit to degrade the RGB image data into RAW data. This process is repeated iteratively until the image pair construction quality indicated by the feedback signal meets the preset requirements. The number of iterations can also be a preset value based on user needs, such as 500 times. After 500 iterations, regardless of the value of the feedback signal, no further iterations are performed. The transformation parameters determined by the policy unit at the 500th iteration are used as the transformation parameters required by the image degradation unit, and the RGB image data obtained by the data processing network is converted into RAW data based on these transformation parameters. In other words, the transformation parameters determined in the last iteration are used as the final transformation parameters. Alternatively, different transformation parameters updated based on the feedback signal during the policy unit search process can be used as the final parameters. For example, when the loss value of the validation dataset is used as the feedback signal, the transformation parameter corresponding to the minimum loss value of the validation dataset can be selected as the final parameter.

[0055] Furthermore, it should be noted that the strategy unit in this application can determine the transformation parameters through networks such as long short-term memory (LSTM) and recurrent neural network (RNN). This application does not make specific limitations here, and any network that can determine the transformation parameters based on AutoML is applicable to this application.

[0056] The data processing method used in this application will be described below. This data processing method can be executed in the aforementioned data processing network, or in a server or a data processing device with data processing capabilities. This application does not specifically limit the executing entity of the data processing method here. The following description uses a server as the executing entity. The server can be referred to... Figure 4 Perform the following steps:

[0057] Step 401: Obtain reference data; wherein the reference data includes: RGB image data and device parameters of the image device.

[0058] Step 402: Use AutoML to determine the multiple conversion parameters required to convert RGB image data into RAW data.

[0059] Step 403: The RGB image data is processed into RAW data according to multiple conversion parameters; the RAW data is matched with the device parameters of the image device.

[0060] This application utilizes AutoML to determine multiple conversion parameters required for converting RGB image data into RAW data, and then converts the RGB image data into RAW data based on these parameters. RAW data obtained in this way improves data processing efficiency, and the RAW data matches the device parameters of the image device, making it more suitable for the device's requirements.

[0061] For example, the server can also determine a search space corresponding to the image device based on device parameters; and then determine multiple transformation parameters from the search space.

[0062] It should be noted that different image devices have different device parameters. This application can adapt a search space that meets the requirements of the image device according to the device parameters of different image devices, and determine the transformation parameters that meet the requirements of the image device within the determined search space. The transformation parameters determined in this way can improve the efficiency of obtaining transformation parameters, and the transformation parameters that meet the requirements of the image device are conducive to better converting RGB image data into RAW data. This application will not specifically describe how to implement the function of the strategy unit in the above data processing network.

[0063] For example, the server can also construct image pairs of RGB image data and RAW data; then input the image pairs into the task processing unit for training to determine the feedback signal; finally, update multiple conversion parameters required to convert RGB image data into RAW data based on the feedback signal; wherein, the task processing unit is used to process video or image data; and the feedback signal is used to indicate the construction quality of the image pairs.

[0064] It should be noted that the feedback signal in this application is used to indicate the construction quality of the image pairs of generated RGB image data and RAW data. Adjusting multiple conversion parameters required for converting RGB image data into RAW data based on the feedback signal can improve the accuracy of the conversion parameters, making the RAW data more suitable for the device parameters of the image device.

[0065] For example, the search space corresponding to the image device includes multiple image processing modules as described in the image degradation unit of the data processing network above. The image processing module includes one or more of the following: a noise addition module, a mosaic addition module, and a brightness adjustment module, which will not be described in detail here.

[0066] For example, the conversion parameters include one or more of the following: noise addition parameters, mosaic parameters, brightness adjustment parameters, gamma parameters, level adjustment parameters, and white balance adjustment parameters, as described in the above data processing network, and will not be repeated here.

[0067] Based on the same concept, such as Figure 4 The image shows a data processing apparatus 500 provided in this application. Exemplarily, the data processing apparatus 500 may be a chip or a chip system. Optionally, in the embodiments of this application, the chip system may be composed of chips, or may include chips and other discrete devices.

[0068] The data processing apparatus 500 may include at least one processor 510, and may also include at least one memory 520 for storing computer programs, program instructions, and / or data. The memory 520 and the processor 510 are coupled. The coupling in this embodiment is an indirect coupling or communication connection between devices, units, or modules, and may be electrical, mechanical, or other forms, used for information exchange between devices, units, or modules. The processor 510 may operate in conjunction with the memory 520. The processor 510 may execute the computer program stored in the memory 520. Optionally, at least one of the at least one memory 520 may be included in the processor 510.

[0069] The data processing device 500 may also include a transceiver 530, through which the data processing device 500 can exchange information with other devices. The transceiver 530 may be a circuit, a bus, a transceiver, or any other device that can be used for information exchange.

[0070] In one possible implementation, the data processing device 500 can be applied to the aforementioned network device. Specifically, the fault analysis device 500 can be the aforementioned network device, or it can be any device capable of supporting the aforementioned network device in implementing any of the above embodiments. The memory 520 stores the necessary computer programs, program instructions, and / or data for implementing the functions of the network device in any of the above embodiments. The processor 510 can execute the computer program stored in the memory 520 to complete the method in any of the above embodiments.

[0071] This application embodiment does not limit the specific connection medium between the transceiver 530, processor 510, and memory 520. This application embodiment... Figure 5 The memory 520, processor 510, and transceiver 530 are connected via a bus, and the bus is in Figure 5 The connections between other components are shown in bold and are for illustrative purposes only, not as limiting information. The bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0072] In the embodiments of this application, the processor may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components, and may implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0073] In the embodiments of this application, the memory can be non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), or it can be volatile memory, such as random-access memory (RAM). The memory can also be any other medium capable of carrying or storing desired program code in the form of instructions or data structures, and accessible by a computer, but is not limited thereto. The memory in the embodiments of this application can also be a circuit or any other device capable of implementing storage functions, used to store computer programs, program instructions, and / or data.

[0074] Based on the above embodiments, this application also provides a readable storage medium storing instructions that, when executed, cause the method performed by the security detection device in any of the above embodiments to be implemented. The readable storage medium may include various media capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory, random access memory, magnetic disk, or optical disk.

[0075] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0076] This application is described with reference to flowchart illustrations and / or block diagrams of the methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0077] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0078] These computer program instructions can also be loaded onto a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable device for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

Claims

1. A data processing method, characterized in that, include: Obtain reference data; wherein, the reference data includes: RGB image data, and device parameters of the image device; AutoML is used to determine the search space corresponding to the image device based on the device parameters, including: ISO and aperture value; Multiple transformation parameters are determined from the search space; The RGB image data is processed into RAW data according to the multiple conversion parameters; the RAW data is matched with the device parameters of the image device.

2. The method according to claim 1, characterized in that, Also includes: Construct image pairs of the RGB image data and the RAW image data; The image is used to train the input task processing unit to determine the feedback signal; The task processing unit is used to process video or image data; the feedback signal is used to indicate the construction quality of the image pair; The multiple conversion parameters required to convert the RGB image data into RAW data are updated based on the feedback signal.

3. The method according to claim 1 or 2, characterized in that, The search space corresponding to the image device includes multiple image processing modules; The image processing module includes one or more of the following: a noise addition module, a mosaic addition module, and a brightness adjustment module.

4. The method according to claim 1 or 2, characterized in that, The conversion parameters include one or more of the following: noise addition parameters, mosaic addition parameters, brightness adjustment parameters, gamma parameters, level adjustment parameters, and white balance adjustment parameters.

5. A data processing system, characterized in that, include: Image degradation unit and strategy unit; The image degradation unit is used to convert RGB image data into RAW data that matches the device parameters of the image device according to multiple conversion parameters output by the strategy unit; The strategy unit is used to determine the search space corresponding to the image device based on the device parameters using AutoML; The plurality of conversion parameters are determined from the search space, the device parameters including: sensitivity and aperture value.

6. The system according to claim 5, characterized in that, The image degradation unit is also used for: Output the image pair of the RGB image data and the RAW data.

7. The system according to claim 6, characterized in that, Also includes: Task processing unit; The task processing unit is used to train on the image pairs of RGB image data and RAW data output by the image degradation unit, determine the feedback signal, and input the feedback signal to the strategy unit; the feedback signal is used to indicate the construction quality of the image pair.

8. The system according to claim 7, characterized in that, The strategy unit is also used for: The system receives the feedback signal output by the task processing unit and adjusts the network parameters of the strategy unit according to the feedback signal. The multiple transformation parameters are updated based on the adjusted network parameters.

9. The system according to any one of claims 5-8, characterized in that, The image degradation unit includes multiple image processing modules; The image processing module includes one or more of the following: a noise addition module, a mosaic addition module, and a brightness adjustment module.

10. The system according to any one of claims 5-8, characterized in that, The conversion parameters include one or more of the following: noise addition parameters, mosaic addition parameters, brightness adjustment parameters, gamma parameters, level adjustment parameters, and white balance adjustment parameters.

11. A data processing apparatus, characterized in that, include: Processor and memory; The memory stores computer programs; The processor is configured to execute a computer program stored in the memory such that the method described in any one of claims 1-4 is performed.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1-4.

Images