CN114077887A - Processing method, device and equipment before point-by-point correction of display screen and storage medium - Google Patents
Processing method, device and equipment before point-by-point correction of display screen and storage medium Download PDFInfo
- Publication number
- CN114077887A CN114077887A CN202111306342.2A CN202111306342A CN114077887A CN 114077887 A CN114077887 A CN 114077887A CN 202111306342 A CN202111306342 A CN 202111306342A CN 114077887 A CN114077887 A CN 114077887A
- Authority
- CN
- China
- Prior art keywords
- pixel image
- image brightness
- array
- point
- display screen
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Processing (AREA)
Abstract
The application discloses a processing method, a device, equipment and a storage medium before point-by-point correction of a display screen, wherein the method comprises the following steps: under different influence factors, acquiring an original pixel image brightness array and a corresponding actual pixel image brightness array after hardware adjustment to generate a training sample set; constructing a convolution neural network model for removing invalid information in pixel image brightness distribution by taking an original pixel image brightness array as input and an actual pixel image brightness array as output; training the convolutional neural network model by utilizing a training sample set; and inputting the pixel image brightness array to be processed into the trained convolutional neural network model for processing, and outputting effective pixel image brightness array information. The convolutional neural network model is constructed and trained, invalid information is deducted from a pixel array captured by a camera in an algorithm mode, tolerance for obtaining proper imaging is further increased, and erection time of a display screen before point-by-point correction is shortened.
Description
Technical Field
The invention relates to the technical field of display, in particular to a processing method, a device, equipment and a storage medium before point-by-point correction of a display screen.
Background
At present, before the conventional point-by-point correction (especially low gray) is performed on the LED or AMOLED display screen, under the condition that the focal length is set, in order to prevent each pixel from being sticky and ensure proper saturation and imaging size, the aperture of the lens, the exposure time (shutter), the distance between the camera and the display screen, the focal length and the like must be adjusted back and forth, and the erection time is very long.
For example, in conventional shooting of low gray useful pixel array brightness information, if the blue saturation is insufficient, the exposure time is increased or the aperture is increased to achieve a proper saturation, but this is likely to cause the blue image size to be too large, and at the same time, the exposure time is decreased or the aperture is increased, so that the exposure time or aperture is adjusted back and forth until the blue pixel is not sticky and the saturation and the image size are both proper. Then, the hardware optical parameters are adjusted repeatedly for both green and red pixels until the green and red pixels are not sticky and the saturation and the imaging size are both proper. Thus, the red, green and blue pixels are not sticky and the respective saturation and image size are within the proper range, resulting in a long setup time.
Therefore, how to solve the problem of long setup time before point-by-point correction of the display screen is a technical problem to be urgently solved by the technical personnel in the field.
Disclosure of Invention
In view of the above, the present invention provides a processing method, an apparatus, a device and a storage medium before point-by-point correction of a display screen, which can increase tolerance for obtaining a proper image before point-by-point correction of the display screen, and reduce a long time for setting up a photographing environment. The specific scheme is as follows:
a processing method before point-by-point correction of a display screen comprises the following steps:
under different influence factors, acquiring an original pixel image brightness array and a corresponding actual pixel image brightness array after hardware adjustment to generate a training sample set;
taking the original pixel image brightness array as input and the actual pixel image brightness array as output, and constructing a convolution neural network model for removing invalid information in pixel image brightness distribution;
training the convolutional neural network model by using the training sample set until the network is converged;
and inputting the pixel image brightness array to be processed into the trained convolutional neural network model for processing, and outputting effective pixel image brightness array information.
Preferably, in the processing method before the point-by-point correction of the display screen provided in the embodiment of the present invention, the influence factor includes any one of an aperture size of a lens, exposure time, a shooting distance between a camera and the display screen, a focal length of the camera, and a bead shape.
Preferably, in the processing method before the point-by-point correction of the display screen provided in the embodiment of the present invention, the training of the convolutional neural network model by using the training sample set includes:
inputting the original pixel image luminance array into the convolutional neural network model;
performing convolution operation on the original pixel image brightness array and the corresponding convolution layers to obtain a pixel brightness characteristic matrix;
performing pooling operation on the pixel brightness characteristic matrix to obtain a target pixel brightness characteristic matrix;
and carrying out nonlinear mapping on the target pixel brightness characteristic matrix through an activation function to obtain the corresponding actual pixel image brightness array.
Preferably, in the processing method before the point-by-point correction of the display screen provided in the embodiment of the present invention, while performing convolution operation, the method further includes:
adjusting the convolution kernel value of the convolution layer for multiple times until the loss function is converged to obtain the optimal convolution kernel value corresponding to the influence factor; the convolution layer has a distribution of convolution kernel values such that a middle value in the matrix is a maximum value and surrounding values surrounding the middle value are smaller than the middle value.
Preferably, in the processing method before the point-by-point correction of the display screen provided in the embodiment of the present invention, before performing convolution operation on the original pixel image luminance array and the corresponding plurality of convolution layers, the method further includes:
expanding the original pixel image brightness array to obtain an expanded first matrix;
correspondingly, performing convolution operation on the original pixel image brightness array and the corresponding convolution layers, including:
and performing convolution operation on the expanded first matrix and the corresponding convolution layers.
Preferably, in the processing method before the point-by-point correction of the display screen provided in the embodiment of the present invention, before performing pooling operation on the pixel brightness feature matrix, the method further includes:
expanding the pixel brightness characteristic matrix to obtain an expanded second matrix;
correspondingly, performing pooling operation on the pixel brightness characteristic matrix, including:
and performing pooling operation on the expanded second matrix.
Preferably, in the processing method before the point-by-point correction of the display screen provided in the embodiment of the present invention, an expression of the activation function is:
and Z represents a numerical value in the target pixel brightness characteristic matrix, and Z represents a numerical value in a pixel brightness pixel image brightness array output by the convolutional neural network model.
The embodiment of the invention also provides a processing device before point-by-point correction of a display screen, which comprises:
the sample generation module is used for acquiring an original pixel image brightness array and a corresponding actual pixel image brightness array after hardware adjustment under different influence factors to generate a training sample set;
the model construction module is used for constructing a convolution neural network model for removing invalid information in pixel image brightness distribution by taking the original pixel image brightness array as input and the actual pixel image brightness array as output;
the model training module is used for training the convolutional neural network model by utilizing the training sample set until the network is converged;
and the processing module is used for inputting the pixel image brightness array to be processed into the trained convolutional neural network model for processing and outputting effective pixel image brightness array information.
The embodiment of the invention also provides processing equipment before point-by-point correction of the display screen, which comprises a processor and a memory, wherein the processing method before point-by-point correction of the display screen is realized when the processor executes a computer program stored in the memory.
The embodiment of the present invention further provides a computer-readable storage medium for storing a computer program, where the computer program, when executed by a processor, implements the processing method before the point-by-point correction of the display screen provided in the embodiment of the present invention.
According to the technical scheme, the processing method before the point-by-point correction of the display screen, provided by the invention, comprises the following steps: under different influence factors, acquiring an original pixel image brightness array and a corresponding actual pixel image brightness array after hardware adjustment to generate a training sample set; constructing a convolution neural network model for removing invalid information in pixel image brightness distribution by taking an original pixel image brightness array as input and an actual pixel image brightness array as output; training the convolutional neural network model by using a training sample set until the network converges; and inputting the pixel image brightness array to be processed into the trained convolutional neural network model for processing, and outputting effective pixel image brightness array information.
According to the processing method before the point-by-point correction of the display screen, provided by the invention, the convolutional neural network model is constructed and trained in a neural network algorithm mode, invalid information is deducted from a pixel array captured by a camera in an algorithm mode, the tolerance for obtaining proper imaging is increased before the point-by-point correction of the display screen is carried out, the lengthy time for setting up a photographing environment is reduced, and an image with high tolerance can be quickly obtained without depending on the traditional hardware optical parameter adjustment.
In addition, the invention also provides a corresponding device, equipment and a computer readable storage medium aiming at the processing method before the point-by-point correction of the display screen, so that the method has higher practicability, and the device, the equipment and the computer readable storage medium have corresponding advantages.
Drawings
In order to more clearly illustrate the embodiments of the present invention or technical solutions in related arts, the drawings used in the description of the embodiments or related arts will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a processing method before point-by-point correction of a display screen according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a processing method of a display screen before point-by-point calibration according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating the relationship between different exposure times and actual pixel image luminance arrays according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating the relationship between different aperture sizes and the actual pixel image luminance array according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a processing device before point-by-point correction of a display screen according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a processing method before point-by-point correction of a display screen, which comprises the following steps as shown in figure 1:
s101, under different influence factors, acquiring an original pixel image brightness array and a corresponding actual pixel image brightness array after hardware adjustment to generate a training sample set;
specifically, the influence factor may include any one of an aperture size of the lens, exposure time, a shooting distance between the camera and the display screen, a focal length of the camera, and a bead form. In practical applications, the display screen may be an LED or AMOLED or other display screen.
The original pixel image brightness array is obtained before hardware adjustment, that is, the original pixel image brightness array without optical parameters being adjusted. The system comprises a red pixel image brightness array, a green pixel image brightness array and a blue pixel image brightness array.
The hardware adjustment refers to adjusting the aperture size of the lens, the exposure time, the shooting distance between the camera and the display screen, the focal length of the camera, the bead shape and the like, and the red, green and blue pixels in the brightness array of the actual pixel image obtained after the hardware adjustment are not sticky, and the individual saturation and the imaging size are in a proper range.
S102, constructing a convolution neural network model for removing invalid information in pixel image brightness distribution by taking an original pixel image brightness array as input and an actual pixel image brightness array as output;
as shown in fig. 2, the original pixel image luminance array with unadjusted optical parameters is processed by an artificial intelligence algorithm (i.e. an algorithm in a convolutional neural network model), and the pixel luminance array with the optical parameters adjusted by hardware can be output.
S103, training the convolutional neural network model by using a training sample set until the network is converged;
s104, inputting the pixel image brightness array to be processed into the trained convolutional neural network model for processing, and outputting effective pixel image brightness array information.
It should be noted that, in the present invention, the artificial intelligence algorithm can replace the adjustment time of the hardware optical parameters before the point-by-point correction of the display screen, and the trained convolutional neural network model can directly output the most appropriate pixel array brightness distribution required before the correction.
In the processing method before the point-by-point correction of the display screen provided by the embodiment of the invention, the convolutional neural network model is constructed and trained by using a neural network algorithm, the invalid information is deducted from the pixel array captured by the camera in an algorithm mode, and then the tolerance for obtaining proper imaging is increased before the point-by-point correction of the display screen is carried out, the lengthy time for setting up a photographing environment is reduced, and an image with higher tolerance can be quickly obtained without depending on the traditional hardware optical parameter adjustment.
Further, in a specific implementation, in the processing method before the point-by-point correction of the display screen provided in the embodiment of the present invention, the step S103 trains the convolutional neural network model by using the training sample set, which may specifically include: firstly, inputting an original pixel image brightness array into a convolution neural network model; then, carrying out convolution operation on the original pixel image brightness array and the corresponding convolution layers to obtain a pixel brightness characteristic matrix; then, performing pooling operation on the pixel brightness characteristic matrix to obtain a target pixel brightness characteristic matrix; and finally, carrying out nonlinear mapping on the target pixel brightness characteristic matrix through an activation function to obtain a corresponding actual pixel image brightness array.
In the training process, the weights of the individual influence factors need to be found in a convolutional neural network manner, and an appropriate convolutional kernel value needs to be trained. Therefore, in practical implementation, the convolution operation performed in the above step may further include: and adjusting the convolution kernel value of the convolution layer for multiple times until the loss function is converged to obtain the optimal convolution kernel value corresponding to the influence factor.
Preferably, in training the network, the distribution of the convolution kernel values of the convolution layer is such that the middle value within the matrix is the maximum value and the surrounding values around the middle value are smaller than the middle value, i.e. the distribution of the convolution kernel values of the convolution layer may be set to be middle around small, for example, when the size of the convolution layer is 5 × 5, the 5 × 5 convolution kernel matrix may be set to:
the middle value of the 5 x 5 convolution kernel matrix is 2, the peripheral values around 2 are 1, and the peripheral values around 1 are 0; the convolution kernel value can be used as the optimal convolution kernel value corresponding to the influence factor.
In addition, in a specific implementation, in the processing method before correcting the display screen point by point according to an embodiment of the present invention, in order to make the arrays correspond to the convolution layers in a one-to-one manner, before performing convolution operation on the original pixel image luminance array and the corresponding convolution layers, the method may further include: expanding the original pixel image brightness array to obtain an expanded first matrix;
correspondingly, convolving the original pixel image luma array with a corresponding plurality of convolution layers may include: and performing convolution operation on the expanded first matrix and the corresponding convolution layers.
Similarly, in a specific implementation, in the processing method before the point-by-point correction of the display screen according to the embodiment of the present invention, in order to make the feature matrix correspond to the numerical value of the pooling layer in a one-to-one correspondence, before performing the pooling operation on the pixel brightness feature matrix, the method may further include: expanding the pixel brightness characteristic matrix to obtain an expanded second matrix;
correspondingly, performing a pooling operation on the pixel brightness feature matrix may include: and performing pooling operation on the expanded second matrix.
In a specific implementation, in the processing method before the point-by-point correction of the display screen provided in the embodiment of the present invention, the expression of the activation function is:
wherein Z represents the value in the target pixel brightness characteristic matrix, and Z represents the value in the pixel brightness pixel image brightness array output by the convolutional neural network model.
It can be understood that, after the numerical values in the target pixel brightness characteristic matrix are substituted into the formula of the activation function, the obtained numerical values are the numerical values of the corresponding positions in the pixel brightness pixel image brightness array output by the convolutional neural network model.
The processing method before the point-by-point correction of the display screen provided by the embodiment of the invention is illustrated by a specific example, which specifically includes the following steps:
step one, shooting an m × n pixel brightness array (for example, blue) of a certain color, wherein the array B is as follows:
step two, expanding the m × n pixel brightness array into an (m +4) × (n +4) matrix, wherein the matrix is as follows:
step three, selecting a proper 5 x 5 convolution kernel:
it should be noted that, in the above example of a suitable 5 × 5 convolution kernel, the different (m +4) × (n +4) matrices in step two are convolved with convolution kernels, and the convolution kernels are all 5 × 5, but the matrix values in the convolution kernels may be different;
and step four, convolving the (m +4) × (n +4) matrix with a convolution kernel to obtain a pixel brightness characteristic matrix m × n, wherein the matrix X is as follows:
step five, expanding the pixel brightness characteristic matrix m × n into a 2m × 2n matrix, wherein the matrix is as follows:
step six, performing pooling operation on the 2m x 2n matrix to obtain a target pixel brightness characteristic matrix m x n, wherein the matrix Y is as follows:
it should be noted that the purpose of pooling is to amplify the feature quantity, and 2 × 2 sampling is performed on the numerical value in step five;
selecting a proper activation function, wherein the activation function is as follows:
step eight, outputting effective pixel array information Zh：
The specific steps of training the optimal convolution kernel value comprise the following steps:
firstly, fixing an input variable and actually measuring an actual m x n brightness data array adjusted by hardware under the condition of the variable;
as shown in fig. 3, 5 exposure times are input to obtain corresponding 5 hardware-adjusted actual m × n luminance data arrays (i.e., actual output matrix Z)1) (ii) a As shown in fig. 4, 5 aperture sizes are input to obtain corresponding 5 hardware-adjusted actual m × n luminance data arrays (i.e., actual output matrix Z2);
second, compute loss functions J1 and J2:
&
wherein p represents the number of samples, ZhThe effective pixel array information Z finally outputted in the above step eight1Representing the actual output matrix, Z, obtained when the exposure time is the influencing factor2Representing the actual output matrix when the aperture size is the influencing factor.
Thirdly, calculating a minimum loss function by adopting the following formula to obtain a proper convolution kernel value;
&
wherein A is1Representing a certain actual input variable, A2Representing another actual input variable, KiRepresenting the input variable A1Corresponding ith convolution kernel, K'iRepresenting the input variable A2Corresponding i-th convolution kernel, Ki+1Representing the input variable A1Corresponding i +1 th convolution kernel, K'i+1Representing the input variable A2The corresponding i +1 th convolution kernel.
It should be noted that, the input impact factors of the above example only consider the aperture size and the exposure time, and in fact, if the two impact factors are not considered completely, the shooting distance, the bead shape, etc. may be added, different convolution kernel weights and residual values are assigned, and training is performed again to find out the optimal convolution kernel value and the optimal residual value corresponding to the individual impact factors, so as to improve the accuracy of the network.
Based on the same inventive concept, the embodiment of the present invention further provides a processing apparatus before point-by-point correction of a display screen, and because the principle of the apparatus for solving the problem is similar to the aforementioned processing method before point-by-point correction of a display screen, the implementation of the apparatus can refer to the implementation of the processing method before point-by-point correction of a display screen, and repeated details are not repeated.
In specific implementation, the processing apparatus before point-by-point correction of the display screen provided in the embodiment of the present invention, as shown in fig. 5, specifically includes:
the sample generation module 11 is configured to obtain an original pixel image brightness array and a corresponding actual pixel image brightness array after hardware adjustment under different influence factors, and generate a training sample set;
the model construction module 12 is used for constructing a convolution neural network model for removing invalid information in pixel image brightness distribution by taking the original pixel image brightness array as input and the actual pixel image brightness array as output;
the model training module 13 is used for training the convolutional neural network model by using a training sample set until the network converges;
and the processing module 14 is used for inputting the pixel image brightness array to be processed into the trained convolutional neural network model for processing, and outputting effective pixel image brightness array information.
In the processing device before the point-by-point correction of the display screen provided by the embodiment of the invention, a convolutional neural network model can be constructed and trained by the interaction of the four modules and by using a neural network algorithm mode, invalid information is deducted from a pixel array captured by a camera in an algorithm mode, and thus the tolerance for obtaining proper imaging is increased before the point-by-point correction of the display screen, the lengthy time for setting up a photographing environment is reduced, and hardware optical parameter adjustment is not needed.
For more specific working processes of the modules, reference may be made to corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.
Correspondingly, the embodiment of the invention also discloses processing equipment before the point-by-point correction of the display screen, which comprises a processor and a memory; the processing method before the point-by-point correction of the display screen disclosed in the foregoing embodiments is implemented when the processor executes the computer program stored in the memory.
For more specific processes of the above method, reference may be made to corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.
Further, the present invention also discloses a computer readable storage medium for storing a computer program; the computer program, when executed by a processor, implements the method of processing before the point-by-point correction of the display screen disclosed above.
For more specific processes of the above method, reference may be made to corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device, the equipment and the storage medium disclosed by the embodiment correspond to the method disclosed by the embodiment, so that the description is relatively simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
To sum up, the processing method before the point-by-point correction of the display screen provided by the embodiment of the invention comprises the following steps: under different influence factors, acquiring an original pixel image brightness array and a corresponding actual pixel image brightness array after hardware adjustment to generate a training sample set; constructing a convolution neural network model for removing invalid information in pixel image brightness distribution by taking an original pixel image brightness array as input and an actual pixel image brightness array as output; training the convolutional neural network model by using a training sample set until the network converges; and inputting the pixel image brightness array to be processed into the trained convolutional neural network model for processing, and outputting effective pixel image brightness array information. The method utilizes a neural network algorithm mode to construct and train a convolutional neural network model, and subtracts invalid information from a pixel array captured by a camera in an algorithm mode, so that tolerance for obtaining proper imaging is increased before point-by-point correction is carried out on a display screen, the lengthy time for setting up a photographing environment is shortened, and an image with high tolerance can be quickly obtained without depending on traditional hardware optical parameter adjustment. In addition, the invention also provides a corresponding device, equipment and a computer readable storage medium aiming at the processing method before the point-by-point correction of the display screen, so that the method has higher practicability, and the device, the equipment and the computer readable storage medium have corresponding advantages.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The processing method, device, equipment and storage medium before the point-by-point correction of the display screen provided by the invention are described in detail, a specific example is applied in the text to explain the principle and the implementation mode of the invention, and the description of the above embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A processing method before point-by-point correction of a display screen is characterized by comprising the following steps:
under different influence factors, acquiring an original pixel image brightness array and a corresponding actual pixel image brightness array after hardware adjustment to generate a training sample set;
taking the original pixel image brightness array as input and the actual pixel image brightness array as output, and constructing a convolution neural network model for removing invalid information in pixel image brightness distribution;
training the convolutional neural network model by using the training sample set until the network is converged;
and inputting the pixel image brightness array to be processed into the trained convolutional neural network model for processing, and outputting effective pixel image brightness array information.
2. The processing method before point-by-point correction of the display screen according to claim 1, wherein the influence factors include any of an aperture size of a lens, an exposure time, a shooting distance between a camera and the display screen, a focal length of the camera, and a bead shape.
3. The processing method before point-by-point correction of the display screen according to claim 2, wherein training the convolutional neural network model by using the training sample set comprises:
inputting the original pixel image luminance array into the convolutional neural network model;
performing convolution operation on the original pixel image brightness array and the corresponding convolution layers to obtain a pixel brightness characteristic matrix;
performing pooling operation on the pixel brightness characteristic matrix to obtain a target pixel brightness characteristic matrix;
and carrying out nonlinear mapping on the target pixel brightness characteristic matrix through an activation function to obtain the corresponding actual pixel image brightness array.
4. The processing method before the point-by-point correction of the display screen according to claim 3, while performing the convolution operation, further comprising:
adjusting the convolution kernel value of the convolution layer for multiple times until the loss function is converged to obtain the optimal convolution kernel value corresponding to the influence factor; the convolution layer has a distribution of convolution kernel values such that a middle value in the matrix is a maximum value and surrounding values surrounding the middle value are smaller than the middle value.
5. The method of claim 3, further comprising, before convolving the original pixel image luminance array with the corresponding plurality of convolution layers:
expanding the original pixel image brightness array to obtain an expanded first matrix;
correspondingly, performing convolution operation on the original pixel image brightness array and the corresponding convolution layers, including:
and performing convolution operation on the expanded first matrix and the corresponding convolution layers.
6. The processing method before the point-by-point correction of the display screen according to claim 3, before the pooling operation of the pixel brightness feature matrix, further comprising:
expanding the pixel brightness characteristic matrix to obtain an expanded second matrix;
correspondingly, performing pooling operation on the pixel brightness characteristic matrix, including:
and performing pooling operation on the expanded second matrix.
7. The processing method before point-by-point correction of the display screen according to claim 3, wherein the expression of the activation function is as follows:
and Z represents a numerical value in the target pixel brightness characteristic matrix, and Z represents a numerical value in a pixel brightness pixel image brightness array output by the convolutional neural network model.
8. A processing device before point-by-point correction of a display screen is characterized by comprising:
the sample generation module is used for acquiring an original pixel image brightness array and a corresponding actual pixel image brightness array after hardware adjustment under different influence factors to generate a training sample set;
the model construction module is used for constructing a convolution neural network model for removing invalid information in pixel image brightness distribution by taking the original pixel image brightness array as input and the actual pixel image brightness array as output;
the model training module is used for training the convolutional neural network model by utilizing the training sample set until the network is converged;
and the processing module is used for inputting the pixel image brightness array to be processed into the trained convolutional neural network model for processing and outputting effective pixel image brightness array information.
9. A processing device before point-by-point correction of a display screen, comprising a processor and a memory, wherein the processor implements a processing method before point-by-point correction of a display screen according to any one of claims 1 to 7 when executing a computer program stored in the memory.
10. A computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements a processing method of a display screen according to any one of claims 1 to 7 before point-by-point correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111306342.2A CN114077887A (en) | 2021-11-05 | 2021-11-05 | Processing method, device and equipment before point-by-point correction of display screen and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111306342.2A CN114077887A (en) | 2021-11-05 | 2021-11-05 | Processing method, device and equipment before point-by-point correction of display screen and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114077887A true CN114077887A (en) | 2022-02-22 |
Family
ID=80283662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111306342.2A Pending CN114077887A (en) | 2021-11-05 | 2021-11-05 | Processing method, device and equipment before point-by-point correction of display screen and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114077887A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116055599A (en) * | 2022-08-19 | 2023-05-02 | 荣耀终端有限公司 | Method for acquiring included angle of folding screen and electronic equipment |
-
2021
- 2021-11-05 CN CN202111306342.2A patent/CN114077887A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116055599A (en) * | 2022-08-19 | 2023-05-02 | 荣耀终端有限公司 | Method for acquiring included angle of folding screen and electronic equipment |
| CN116055599B (en) * | 2022-08-19 | 2023-11-24 | 荣耀终端有限公司 | Method for acquiring included angle of folding screen and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200151858A1 (en) | Image contrast enhancement method and device, and storage medium | |
| CN109644230B (en) | Image processing method, image processing apparatus, image pickup apparatus, and storage medium | |
| JP7439145B2 (en) | Image processing method, image processing device, image processing system, learned weight generation method, and program | |
| US11508038B2 (en) | Image processing method, storage medium, image processing apparatus, learned model manufacturing method, and image processing system | |
| KR20040073378A (en) | Vignetting compensation | |
| CN102714737A (en) | Image processing device and image capture apparatus using same | |
| US20120262610A1 (en) | Pixel Information Reproduction Using Neural Networks | |
| CN113313661A (en) | Image fusion method and device, electronic equipment and computer readable storage medium | |
| CN114077887A (en) | Processing method, device and equipment before point-by-point correction of display screen and storage medium | |
| JP2021140663A (en) | Image processing method, image processing device, image processing program, and recording medium | |
| CN112243119B (en) | White balance processing method and device, electronic equipment and storage medium | |
| JP2012204856A (en) | Image processing method and camera module | |
| CN116029938B (en) | Image flat field correction method and device based on linear fitting | |
| CN117156289A (en) | Color style correction method, system, electronic device, storage medium and chip | |
| CN111815720B (en) | Image processing method and device, readable storage medium and electronic equipment | |
| CN116156333A (en) | Camera imaging optimization method and system based on complex illumination environment | |
| CN113038025B (en) | Image processing method, terminal and storage medium | |
| KR100991146B1 (en) | Satelite image correction method and apparatus using neural network | |
| JP2020030569A (en) | Image processing method, image processing device, imaging device, lens device, program, and storage medium | |
| CN111064860A (en) | Image correction method, image correction device and electronic equipment | |
| CN110827362A (en) | Luminosity calibration method based on polynomial camera response function and vignetting effect compensation | |
| JP2021114186A (en) | Image processing device, image processing method, and program | |
| CN114187185A (en) | Data processing method, system and device | |
| WO2011076975A1 (en) | Filter setup learning for binary sensor | |
| JP2021189929A (en) | Image processing method, program, image processing device and image processing system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Unit 801-A, No. 37-3, Banshangshe, Building 3, Xinke Plaza, Xiamen Torch High-tech Zone, Xiamen, Fujian 361000 Applicant after: Xiamen Lingyang Huaxin Technology Co.,Ltd. Address before: 203-91, South Building, torch Plaza, No. 56-58, torch garden, torch hi tech Zone, Huli District, Xiamen City, Fujian Province, 361000 Applicant before: XIAMEN LINGYANG HUAXIN TECHNOLOGY CO.,LTD. |