CN112801216B - Wavelength compensation method and device, computer equipment and storage medium - Google Patents
Wavelength compensation method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN112801216B CN112801216B CN202110288386.0A CN202110288386A CN112801216B CN 112801216 B CN112801216 B CN 112801216B CN 202110288386 A CN202110288386 A CN 202110288386A CN 112801216 B CN112801216 B CN 112801216B
- Authority
- CN
- China
- Prior art keywords
- wavelength
- sample
- value
- tristimulus values
- prediction model
- 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.)
- Active
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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Spectrometry And Color Measurement (AREA)
Abstract
The application relates to a wavelength compensation method, a wavelength compensation device, a computer device and a storage medium. The method comprises the steps of obtaining a tristimulus value corresponding to a wavelength to be compensated in a display screen, inputting the tristimulus value into a wavelength prediction model, carrying out conversion processing from the tristimulus value to the wavelength on the input tristimulus value through the wavelength prediction model to obtain a corresponding predicted wavelength, and carrying out wavelength compensation on the wavelength to be compensated according to the predicted wavelength. Compared with the traditional mode of compensating by selecting wavelengths with different light transmittance by workers, the scheme utilizes the wavelength prediction model to predict the wavelength of the collected tristimulus values of the display screen, so that the wavelength most suitable for the collected tristimulus values is obtained, the wavelength is compensated, the problem of unmatched existing wavelength is solved, and the effect of improving the compensation efficiency of wavelength compensation is realized.
Description
Technical Field
The present application relates to the field of optical technologies, and in particular, to a wavelength compensation method and apparatus, a computer device, and a storage medium.
Background
An LED display screen is a flat panel display, can be generally used for displaying characters, images and the like, and is a display device commonly used at present. In the production process of the LED display screen, the LED display screen needs to be corrected, and in the correction process, a camera is usually used for correction, but due to process problems, a light filter in the camera has a problem of irregular wavelength light transmittance, and corresponding light transmittance of a charge coupled device in the camera is also irregular under different wavelengths, so that in the use process, a problem of mismatch between an actual wavelength and a detected wavelength is generated, and therefore in the chromaticity correction process of the LED display screen, wavelength difference needs to be compensated. At present, the mode of compensating the wavelength difference is usually to compensate by selecting the wavelengths with different light transmittances by workers, however, the problem of complicated steps exists when the wavelength is compensated by the mode, and the efficiency is low.
Therefore, the current wavelength compensation method has the defect of low efficiency.
Disclosure of Invention
In view of the above, it is desirable to provide a wavelength compensation method, apparatus, computer device, and storage medium capable of improving the efficiency of wavelength compensation in view of the above technical problems.
A method of wavelength compensation, the method comprising:
acquiring a tristimulus value corresponding to a wavelength to be compensated in a display screen;
inputting the tristimulus values into a wavelength prediction model, and performing conversion processing from the tristimulus values to the wavelengths on the input tristimulus values through the wavelength prediction model to obtain corresponding predicted wavelengths; the wavelength prediction model is obtained by training according to the tristimulus values of the multiple samples and the known wavelengths of the samples corresponding to the tristimulus values of the multiple samples;
and according to the predicted wavelength, performing wavelength compensation on the wavelength to be compensated.
In one embodiment, before inputting the tristimulus values to be compensated into the wavelength prediction model, the method further includes:
acquiring the tristimulus values of a preset number of samples acquired according to a display screen under various conditions and the known wavelengths of the preset number of samples; the tristimulus values of the samples in the preset number correspond to the known wavelengths of the samples in the preset number one by one;
normalizing the preset number of sample tristimulus values to obtain a preset number of normalized sample tristimulus values;
according to the normalized sample tristimulus values, acquiring corresponding sample predicted wavelengths through a wavelength prediction model to be trained;
and updating the wavelength prediction model to be trained according to the sample predicted wavelength and the error value corresponding to the known wavelength of the sample until the error value is within a preset value range, so as to obtain the wavelength prediction model.
In one embodiment, the obtaining of the preset number of sample tristimulus values collected according to the display screen under the multiple conditions includes:
acquiring the preset number of measurement tristimulus values which are sent by the light ray acquisition equipment and acquired by the display screen under the various conditions;
for each measured tristimulus value, converting the measured tristimulus value into corresponding red primary color stimulus value, green primary color stimulus value and blue primary color stimulus value to obtain the sample tristimulus value;
and obtaining the sample tristimulus values of the preset number according to the plurality of sample tristimulus values.
In one embodiment, the normalizing the preset number of sample tristimulus values to obtain a preset number of normalized sample tristimulus values includes:
respectively normalizing the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the sample tristimulus values aiming at each sample tristimulus value to obtain normalized sample tristimulus values; the numerical values of the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the normalized sample tristimulus values are all larger than 0 and less than or equal to 1;
and obtaining the normalized sample tristimulus values of the preset number according to the plurality of normalized sample tristimulus values.
In one embodiment, the wavelength prediction model to be trained is a BP neural network prediction model to be trained, and the BP neural network prediction model to be trained comprises an input layer, a hidden layer and an output layer;
the obtaining of the corresponding sample prediction wavelength through the wavelength prediction model to be trained according to the normalized sample tristimulus value comprises the following steps:
aiming at each normalized sample tristimulus value, acquiring an output result corresponding to the hidden layer according to a red primary color stimulus value, a green primary color stimulus value and a blue primary color stimulus value in the normalized sample tristimulus values and a first weight of a BP neural network prediction model to be trained; the first weight represents the corresponding weight from the input layer to the hidden layer;
obtaining a sample prediction wavelength corresponding to the normalized sample tristimulus value according to the output result and a second weight of the BP neural network prediction model to be trained; the second weight represents the weight from the hidden layer to the output layer;
the step of updating the wavelength prediction model to be trained according to the error value corresponding to the sample predicted wavelength and the known wavelength of the sample until the error value is within a preset value range to obtain the wavelength prediction model comprises the following steps:
obtaining an error value between the sample predicted wavelength and a sample known wavelength corresponding to the normalized tristimulus value;
judging whether the error value is smaller than or equal to a preset value;
if not, updating the first weight and the second weight according to the error value and preset learning efficiency, returning to the step of obtaining an output result corresponding to the hidden layer according to the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the normalized sample tristimulus values and the first weight of the BP neural network prediction model to be trained, and executing next model training until the number of model training times reaches preset times;
if so, finishing training the BP neural network prediction model to be trained to obtain the wavelength prediction model.
In one embodiment, the obtaining an output result corresponding to the hidden layer according to the red primary color stimulus value, the green primary color stimulus value, the blue primary color stimulus value in the normalized sample tristimulus values and the first weight of the BP neural network prediction model to be trained includes:
obtaining a first product of a red primary color stimulus value and the first weight, a second product of a green primary color stimulus value and the first weight, and a third product of a blue primary color stimulus value and the first weight in the normalized sample tristimulus values;
and acquiring an output result corresponding to the hidden layer according to the first product, the second product and the third product.
In one embodiment, the updating the first weight and the second weight according to the error value and a preset learning efficiency includes:
acquiring a first adjusting value according to the partial derivative of the error value to the first weight and the preset learning efficiency;
acquiring a second adjusting value according to the partial derivative of the error value to the second weight and the preset learning efficiency;
acquiring the sum of the first weight and the first adjusting value to obtain an updated first weight;
and acquiring the sum of the second weight and the second adjusting value to obtain the updated second weight.
A wavelength compensation device, the device comprising:
the acquisition module is used for acquiring the tristimulus values corresponding to the wavelengths to be compensated in the display screen;
the prediction module is used for inputting the tristimulus values into a wavelength prediction model, and performing conversion processing from the tristimulus values to the wavelengths on the input tristimulus values through the wavelength prediction model to obtain corresponding predicted wavelengths; the wavelength prediction model is obtained by training according to the tristimulus values of the multiple samples and the known wavelengths of the samples corresponding to the tristimulus values of the multiple samples;
and the compensation module is used for carrying out wavelength compensation on the wavelength to be compensated according to the predicted wavelength.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the method described above when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
According to the wavelength compensation method, the device, the computer equipment and the storage medium, the tristimulus values corresponding to the wavelength to be compensated in the display screen are obtained, the tristimulus values are input into the wavelength prediction model, the input tristimulus values are subjected to conversion processing from the tristimulus values to the wavelength through the wavelength prediction model, the corresponding predicted wavelength is obtained, and the wavelength compensation is carried out on the wavelength to be compensated according to the predicted wavelength. Compared with the traditional mode of compensating by selecting wavelengths with different light transmittance by workers, the scheme utilizes the wavelength prediction model to predict the wavelength of the collected tristimulus values of the display screen, so that the wavelength most suitable for the collected tristimulus values is obtained, the wavelength is compensated, the problem of unmatched existing wavelength is solved, and the effect of improving the compensation efficiency of wavelength compensation is realized.
Drawings
FIG. 1 is a diagram of an exemplary wavelength compensation method;
FIG. 2 is a schematic flow chart of a wavelength compensation method according to an embodiment;
FIG. 3 is a schematic diagram of a wavelength prediction model according to an embodiment;
FIG. 4 is a block diagram showing the structure of a wavelength compensation device according to an embodiment;
FIG. 5 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The wavelength compensation method provided by the application can be applied to the application environment as shown in fig. 1. Wherein the server 102 may be in communication with the capture device 104, and the capture device 104 may be connected to a display screen. The acquisition device 104 may acquire related display parameters of a display screen connected thereto, for example, a tristimulus value of the display screen, the acquisition device 104 may transmit the acquired tristimulus value to the server 102, the server 102 may input the acquired tristimulus value into the wavelength prediction model, and convert the tristimulus value into a predicted wavelength through the wavelength prediction model, so that the server 104 may perform wavelength compensation on a wavelength to be compensated with the acquired predicted wavelength, that is, compensate for a wavelength difference problem caused by wavelength mismatch. The server 102 may be implemented by an independent server or a server cluster composed of a plurality of servers; the capture device 104 may be various cameras, luminance meters (e.g., CS 2000), and the like.
In one embodiment, as shown in fig. 2, a wavelength compensation method is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:
step S202, acquiring tristimulus values corresponding to the wavelengths to be compensated in the display screen.
The display screen may be a screen body for displaying information such as images and characters, for example, an LED display screen; the LED display screen is an electronic display screen formed by LED lattices, the display content forms of the screen, such as characters, animations, pictures and videos, are changed in time by the red and green lamp beads which are turned on and off, and the display control of the components is carried out through a modular structure. Mainly comprises a display module, a control system and a power supply system. The display module is a screen consisting of LED lamp dot arrays and emits light; the control system realizes the conversion of the content displayed on the screen under the on-off condition in the regulation area; the power supply system converts the input voltage and current to meet the requirement of the display screen. The tristimulus values are the amounts of the three primary color stimuli required to achieve color matching with the light to be measured in a tristimulus system. Wherein three colors may include red, green and blue, and three colors among the tristimulus values may be represented by X (red primary color stimulus amount), Y (green primary color stimulus amount) and Z (blue primary color stimulus amount).
When the chromaticity of the LED screen is corrected, a camera is used for correction, but due to process problems, the actual wavelength may not match the detected wavelength, so the wavelength to be compensated may be the detected wavelength of the LED screen detected by a light collecting device, such as a camera, and the like, and the server 102 needs to perform wavelength compensation on the detected wavelength, so as to match the actual wavelength. For example, the server 102 may obtain a tristimulus value corresponding to a wavelength to be compensated, the tristimulus value may be acquired by the light collecting device to obtain a tristimulus value corresponding to the LED screen, and the light collecting device may send the tristimulus value to the server 102, so that the server 102 may obtain the tristimulus value corresponding to the LED screen that needs to be wavelength compensated.
Step S204, inputting the tristimulus values into a wavelength prediction model, and performing conversion processing from the tristimulus values to the wavelengths on the input tristimulus values through the wavelength prediction model to obtain corresponding predicted wavelengths; the wavelength prediction model is obtained by training according to the tri-stimulus values of the multiple samples and the known wavelengths of the samples corresponding to the tri-stimulus values of the multiple samples.
The wavelength prediction model may be a neural network model for wavelength prediction, for example, a trained BP (back propagation) neural network model, the BP neural network is a multi-layer feedforward network trained according to error back propagation (error back propagation for short), the algorithm is called BP algorithm, and the basic idea is a gradient descent method, which uses a gradient search technique to minimize the mean square error between the actual output value and the expected output value of the network. In this embodiment, the wavelength prediction model may include a plurality of functions, and the server 102 may input the obtained tristimulus values into the wavelength prediction model, and convert the tristimulus values into corresponding wavelengths by using the wavelength prediction model, so as to obtain the predicted wavelengths. For example, multiple functions may be included in the wavelength prediction model, and server 102 may convert the tristimulus values into corresponding predicted wavelengths using the multiple functions in the wavelength prediction model. For example, the server 102 may train the wavelength prediction model to be trained by using a plurality of sample tristimulus values acquired under a plurality of display screen conditions and the known sample wavelengths corresponding to the sample tristimulus values, so as to obtain the wavelength prediction model for prediction.
And S206, performing wavelength compensation on the wavelength to be compensated according to the predicted wavelength.
The predicted wavelength may be wavelength information obtained by the server 102 by converting the tristimulus values through the wavelength prediction model by using the collected tristimulus values corresponding to the display screen, and the server 102 may use the predicted wavelength as the compensation wavelength of the display screen, so that the wavelength compensation may be performed on the wavelength to be compensated corresponding to the display screen, and the problem of wavelength mismatch is avoided. For example, if the server 102 detects that the wavelength to be compensated corresponding to the display screen acquired by the camera is not matched with the predicted wavelength, the wavelength to be compensated may be adjusted according to the predicted wavelength; the server 102 may also directly use the predicted wavelength as the wavelength corresponding to the display screen. So that the server 102 can perform chromaticity correction on the display screen according to the wavelength.
According to the wavelength compensation method, the tristimulus values corresponding to the wavelengths to be compensated in the display screen are obtained, the tristimulus values are input into the wavelength prediction model, the input tristimulus values are converted from the tristimulus values to the wavelengths through the wavelength prediction model, the corresponding predicted wavelengths are obtained, and the wavelengths to be compensated are subjected to wavelength compensation according to the predicted wavelengths. Compared with the traditional mode of compensating by selecting wavelengths with different light transmittance by workers, the scheme utilizes the wavelength prediction model to predict the wavelength of the collected tristimulus values of the display screen, so that the wavelength most suitable for the collected tristimulus values is obtained, the wavelength is compensated, the problem of unmatched existing wavelength is solved, and the effect of improving the compensation efficiency of wavelength compensation is realized.
In one embodiment, before inputting the tristimulus values to be compensated into the wavelength prediction model, the method further includes: acquiring the tristimulus values of a preset number of samples acquired according to a display screen under various conditions and the known wavelengths of the preset number of samples; the tristimulus values of the samples in the preset number correspond to the known wavelengths of the samples in the preset number one by one; normalizing the preset number of sample tristimulus values to obtain the preset number of normalized sample tristimulus values; according to the normalized sample tristimulus values, acquiring corresponding sample predicted wavelengths through a wavelength prediction model to be trained; and updating the wavelength prediction model to be trained according to the sample predicted wavelength and the error value corresponding to the known wavelength of the sample until the error value is within a preset numerical range to obtain the wavelength prediction model.
In this embodiment, the server 102 may perform conversion processing on the tristimulus values corresponding to the display screen through the wavelength prediction model to obtain the predicted wavelength corresponding to the tristimulus values. The server 102 may obtain the wavelength prediction model through training, for example, the server 102 may obtain a preset number of sample tristimulus values, which may be at least 20 groups of tristimulus values, and the preset number of tristimulus values may be tristimulus values corresponding to a display screen acquired by the acquisition device 104 under various conditions, that is, under various scenes. The server 102 may further obtain a preset number of sample known wavelengths, for example, the known wavelengths of the samples corresponding to the display screen are collected by the collecting device 104, where the number of the sample known wavelengths may be a preset number, and the preset number of the sample known wavelengths corresponds to the tristimulus values of the preset number of samples one to one. In one embodiment, the collection device 104 may be a light collection device such as a luminance meter, including a model CS2000 luminance meter, or the like. The luminance meter can collect the sample tristimulus values corresponding to the display screen and the known wavelengths of the samples corresponding to the tristimulus values of the samples under various conditions, for example, the sample tristimulus values corresponding to the display screen and the known wavelengths of the samples corresponding to the tristimulus values of the samples under various scenes, wherein the various scenes can include different display luminance, different display colors and the like. The tristimulus values may be composed of a red primary color stimulus value, a blue primary color stimulus value and a green primary color stimulus value, and the server 102 may obtain a preset number of measurement tristimulus values, which are sent by the luminance meter and acquired by the display screen under various conditions; and the known wavelength of the sample corresponding to the preset number of the measured tristimulus values sent by the luminance meter can be obtained; for each measured tristimulus value, the server 102 may convert the measured tristimulus value into a corresponding red primary color stimulus value, green primary color stimulus value, and blue primary color stimulus value to obtain a sample tristimulus value, e.g., the server 102 may represent a set of tristimulus values as [ X1; Y1; Z1], where X may represent a red primary color stimulus value, Y may represent a green primary color stimulus value, and Z may represent a blue primary color stimulus value; then the sets of tristimulus values may be expressed as: [ X1X 2 X3... cnt, [ Y1Y 2 Y3... Yn ], [ Z1Z 2 Z3... Zn ]; namely, a group of tristimulus values comprises a red primary color stimulus value, a green primary color stimulus value and a blue primary color stimulus value; and, each set of the above-mentioned sample tristimulus values can correspond to a sample known wavelength L; further, the server 102 may obtain a preset number of sample tristimulus values and a preset number of sample known wavelengths according to the plurality of sample tristimulus values and the sample known wavelengths corresponding to the respective sample tristimulus values. The luminance meter may be a split-light luminance meter, and may be used to measure parameters such as luminance of a display screen, for example, cs2000, and the display screen may be an LED screen body.
Since the activation function used in training is
That is, y is 0 to 1, so that before training, the server 102 may normalize the tristimulus values of the sample of the preset number, thereby obtaining the tristimulus values of the sample of the preset number normalized to 0 to 1; the server 102 may further input the predetermined number of normalized sample tristimulus values into the wavelength prediction model to be trained, so that the corresponding sample predicted wavelength may be obtained by the wavelength prediction model to be trained, and the server 102 may update the wavelength prediction model to be trained according to an error value between the sample predicted wavelength and the known wavelength of the sample until the error value is within a predetermined numerical range, thereby obtaining the wavelength prediction model. For example, the server 102 may update a plurality of weights in the wavelength model to be trained, such that the error value may be gradually decreased after each training. The wavelength prediction model to be trained may be a BP neural network prediction model. The server 102 may obtain the predicted wavelength of the tristimulus value corresponding to the display screen by using the trained wavelength prediction model.
Through the embodiment, the server 102 may train the wavelength prediction model to be trained, so as to predict the wavelength corresponding to the tristimulus value of the display screen by using the trained wavelength prediction model, thereby achieving an effect of improving the compensation efficiency of wavelength compensation.
In one embodiment, normalizing a preset number of sample tristimulus values to obtain a preset number of normalized sample tristimulus values comprises: respectively normalizing the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the sample tristimulus values aiming at each sample tristimulus value to obtain normalized sample tristimulus values; the numerical values of the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the normalized sample tristimulus values are all larger than 0 and less than or equal to 1; and obtaining a preset number of normalized sample tristimulus values according to the plurality of normalized sample tristimulus values.
In this embodiment, the server 102 may normalize the obtained tristimulus values of the preset number of samples before training the wavelength prediction model to be trained, for example, the normalization may be performed on the tristimulus values of the respective samples by using Min-Max normalization, that is, the tristimulus values are normalized to 0 to 1, and the activation function used in the training process is
Y is 0 to 1, so the tristimulus values need to be normalized to between 0 and 1. The server 102 may normalize each of the red primary color stimulus values, the green primary color stimulus values, and the blue primary color stimulus values in the sample tristimulus values, respectively, to obtain normalized sample tristimulus values. Wherein, the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the normalized sample tristimulus values are all numerical values which are more than 0 and less than or equal to 1. After the server 102 normalizes the plurality of sample tristimulus values, the predetermined number of normalized sample tristimulus values may be obtained.
Through the embodiment, the server 102 can normalize the sample tristimulus values, so that the server 102 can train the wavelength prediction model to be trained by using the normalized sample tristimulus values, the training effect is improved, the trained wavelength prediction model can predict the wavelength corresponding to the tristimulus values of the display screen, and the effect of improving the compensation efficiency of wavelength compensation is realized.
In one embodiment, obtaining the corresponding sample predicted wavelength through a wavelength prediction model to be trained according to the normalized sample tristimulus value includes: aiming at each normalized sample tristimulus value, acquiring an output result corresponding to a hidden layer according to a red primary color stimulus value, a green primary color stimulus value, a blue primary color stimulus value and a first weight of a BP neural network prediction model to be trained in the normalized sample tristimulus values; the first weight represents the corresponding weight from the input layer to the hidden layer; acquiring a sample prediction wavelength corresponding to the normalized sample tristimulus value according to the output result and the second weight of the BP neural network prediction model to be trained; the second weight represents the weight corresponding to the hidden layer to the output layer.
In this embodiment, the wavelength prediction model to be trained may be a BP neural network prediction model to be trained, and the BP neural network prediction model to be trained includes an input layer, a hidden layer, and an output layer. Wherein the input layer may be a layer for inputting the normalized tristimulus values; the hidden layer can be a layer which carries out first numerical value conversion on the tristimulus values; the output layer may be a layer that performs second conversion on the conversion result of the tristimulus values by the hidden layer to obtain and output a predicted wavelength. The server 102 may train the BP neural network prediction model to be trained by using the sample tristimulus values, for example, the server 102 may set a preset training time threshold to 500 times, and set the learning efficiency to be more than 75%; the BP neural network prediction model to be trained may include input layers of three neurons, each input layer corresponds to one of the sample tristimulus values, and each time training is performed, the server 102 may input a red primary color stimulus value, a green primary color stimulus value, and a blue primary color stimulus value in a group of sample tristimulus values into each neuron in the input layers, and convert and input the red primary color stimulus value, the green primary color stimulus value, and the blue primary color stimulus value into each neuron in the hidden layer, respectively, according to the first weight, to obtain an output result corresponding to the hidden layer. The hidden layer may include 9 neurons, and the hidden layer may be set with reference to a neuron formula of the hidden layer, for example:
(ii) a Wherein N is the number of input layers and L is an output layer; the first weight may represent a weight corresponding to the input layer to the hidden layer, and the weight may be updated in the training process. Specifically, the server 102 outputs the sample tristimulus valuesThe function for the in-layer transition to the hidden layer may be:
(ii) a Wherein Xi, Yi and Zi respectively represent a red primary color stimulus value, a green primary color stimulus value and a blue primary color stimulus value in a group of tristimulus values; i =3, i represents the number of tristimulus values in a set, and Si is the weight from the input layer to the hidden layer, i.e. the first weight; h (x) may be the output result described above.
The server 102 may obtain the sample predicted wavelength corresponding to the normalized sample tristimulus value by using the output result and the second weight or one of the weights in the BP neural network prediction model to be trained, where the second weight represents the weight corresponding to the hidden layer to the output layer. For example, the server 102 may obtain the sample predicted wavelength for the output layer using the following formula:
(ii) a Wherein n =9 in the formula, n in the formula may represent the number of neurons of the hidden layer, λ (i) represents the sample predicted wavelength, Oi is a weight from the hidden layer to the output layer, i.e. the above second weight, and the second weight may be updated in the training process.
In addition, in an embodiment, the server 102 may update the wavelength prediction model to be trained according to the sample predicted wavelength obtained by each training, for example, in an embodiment, update the wavelength prediction model to be trained according to the sample predicted wavelength and an error value corresponding to the known wavelength of the sample until the error value is within a preset value range, so as to obtain the wavelength prediction model, including: obtaining an error value between the sample predicted wavelength and the known wavelength of the sample corresponding to the normalized tristimulus value; judging whether the error value is less than or equal to a preset value; if not, updating the first weight and the second weight according to the error value and preset learning efficiency, returning to the step of obtaining an output result corresponding to the hidden layer according to the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the normalized sample tristimulus values and the first weight of the BP neural network prediction model to be trained, and executing next model training until the number of model training times reaches the preset number of times; if so, finishing training the BP neural network prediction model to be trained to obtain a wavelength prediction model.
In this embodiment, the server 102 may obtain an error value between the predicted wavelength of the sample and the known wavelength of the sample according to the predicted wavelength of the sample, for example, an obtaining formula of the error value may be:
(ii) a Where λ (i) represents the predicted wavelength of the sample, λ t (i) represents the known wavelength of the sample, and Ut is the total difference, i.e., the above-mentioned error value. The server 102 may use a back propagation algorithm to weight the weights between the layers according to the error values, for example, the server 102 may determine the error values, if the error values are greater than a preset value, it indicates that the prediction effect of the wavelength prediction model to be trained is not ideal, at this time, the server 102 may update the first weight and the second weight according to the error values and the preset learning efficiency, and return to the step of obtaining the output result corresponding to the hidden layer according to the first weight of the neural network prediction model to be trained and the red primary color stimulus value, the green primary color stimulus value, and the blue primary color stimulus value in the normalized sample tristimulus values, and the first weight of the BP neural network prediction model to be trained, and continue training the model from the step by using the updated first weight and the updated second weight, including steps of inputting the sample, outputting the predicted wavelength, determining the error values, and the like, thereby realizing the next model training. The error value may be determined in each training, if the error value is not less than or equal to the preset value when the error value is determined in the multiple model training, the server 102 may update the first weight and the second weight in the model for multiple times, and continue to train the model after each update until the number of times of training reaches a preset training threshold, for example, 500 times, if the error value is not less than or equal to the preset value, the server 102 may stop the training, and if the error value is less than or equal to the preset value in the training within the preset training threshold, the server 102 may end the training。
When the error value is less than or equal to the preset value, the server 102 may determine that the wavelength prediction model to be trained is successfully trained, and obtain the wavelength prediction model according to the BP neural network prediction model to be trained when training is finished.
Through the embodiment, the server 102 can continuously update each weight in the wavelength prediction model to be trained by using the tristimulus values, so that the wavelength prediction model can be obtained, and the wavelength prediction model is used for performing wavelength compensation on the display screen, thereby improving the compensation efficiency of the wavelength compensation.
In one embodiment, obtaining an output result corresponding to a hidden layer according to a red primary color stimulus value, a green primary color stimulus value, a blue primary color stimulus value in the normalized sample tristimulus values and a first weight of a BP neural network prediction model to be trained includes: obtaining a first product of a red primary color stimulus value and a first weight, a second product of a green primary color stimulus value and the first weight, and a third product of a blue primary color stimulus value and the first weight in the normalized sample tristimulus values; and acquiring an output result corresponding to the hidden layer according to the first product, the second product and the third product.
In this embodiment, the server 102 may obtain an output result from the input layer to the hidden layer by using each primary color stimulus value in the normalized sample tristimulus values and the first weight in the BP neural network prediction model to be trained. For example, the server 102 may obtain a first product of the red primary color stimulus value and the first weight, a second product of the green primary color stimulus value and the first weight, and a third product of the blue primary color stimulus value and the first weight in the normalized sample tristimulus values, and obtain an output result corresponding to the hidden layer according to the first product, the second product, and the third product. Specifically, the server 102 may obtain the output result by using the following function:
(ii) a Wherein Xi, Yi and Zi respectively represent a red primary color stimulus value, a green primary color stimulus value and a blue primary color stimulus value in a group of tristimulus values; i =3, and the number of the channels is increased,i represents the number of a set of tristimulus values, and Si is the weight from the input layer to the hidden layer, namely the first weight; h (x) may be the output result described above.
Through this embodiment, the server 102 may train the wavelength prediction model to be trained by using each primary color stimulus value and the first weight in the normalized sample tristimulus values, so that the trained wavelength prediction model may be used to perform wavelength compensation, thereby improving the efficiency of wavelength compensation.
In one embodiment, updating the first weight and the second weight according to the error value and the preset learning efficiency includes: acquiring a first adjusting value according to a partial derivative of the error value to the first weight and a preset learning efficiency; acquiring a second adjustment value according to the partial derivative of the error value to the second weight and the preset learning efficiency; acquiring the sum of the first weight and the first adjusting value to obtain an updated first weight; and acquiring the sum of the second weight and the second adjusting value to obtain the updated second weight.
In this embodiment, the server 102 may update the first weight and the second weight in the wavelength prediction model to be trained when the error value between the predicted wavelength of the sample and the known wavelength of the sample is greater than a preset value, for example, the server 102 may obtain a first adjustment value according to a partial derivative of the error value to the first weight and the preset learning efficiency, and obtain a second adjustment value according to a partial derivative of the error value to the second weight and the preset learning efficiency, and the server 102 may further obtain a sum of the first weight and the first adjustment value, so as to obtain the updated first weight, and obtain a sum of the second weight and the second adjustment value, so as to obtain the updated second weight. Specifically, the server 102 may update the first weight and the second weight using the following function:
(ii) a Wherein Si isTThe new input layer to hidden layer weight, i.e. the updated first weight, Si is the first weight before updating,
is a first adjusted value, and
(ii) a Wherein Ut is an error value, η is the preset learning efficiency;
(ii) a Wherein QiTFor the new hidden layer to output layer weight, i.e. the second weight described above,
is a second adjusted value, and
(ii) a In addition, it should be noted that the original weights of the first weight and the second weight, i.e., the weights during the first training, may be randomly assigned.
Through the embodiment, the server 102 may update each weight by using the error value and the learning efficiency, so as to obtain the trained wavelength prediction model, and perform wavelength compensation by using the wavelength prediction model, thereby improving the efficiency of wavelength compensation.
In one embodiment, as shown in fig. 3, fig. 3 is a schematic structural diagram of a wavelength prediction model in one embodiment. Based on the structure diagram, the wavelength compensation method can comprise the following steps:
the server 102 may obtain tristimulus values corresponding to the display screen sent by the acquisition device 104, where the tristimulus values include a red primary color stimulus value, a green primary color stimulus value, and a blue primary color stimulus value; the server 102 may input the red primary color stimulus value, the green primary color stimulus value, and the blue primary color stimulus value in the normalized tristimulus values into the wavelength prediction model, for example, the server 102 may input each of the primary color stimulus values into each of the neurons in the input layer as shown in fig. 3, the server 102 may convert data in the input layer into the hidden layer by using the wavelength prediction model, for example, map each of the primary color stimulus values into each of the neurons in the hidden layer by using the first weight Si, in the hidden layer, the server 102 may obtain an output result by using an h (x) function, and may output the output result into the output layer by combining the second weight Oi to obtain a predicted wavelength λ (i), thereby implementing the wavelength prediction, and the server 102 may perform the wavelength compensation by using the predicted wavelength. The wavelength prediction model can be obtained by training according to the tristimulus values of the multiple samples and the known wavelengths of the samples corresponding to the tristimulus values of the multiple samples.
Through the embodiment, the server 102 can perform wavelength prediction on the collected tristimulus values of the display screen by using the wavelength prediction model, so that the wavelength most suitable for the collected tristimulus values is obtained, the wavelength is compensated, the problem that the existing wavelength is not matched is solved, and the effect of improving the compensation efficiency of wavelength compensation is realized.
It should be understood that, although the steps in the flowchart of fig. 2 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in fig. 2 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.
In one embodiment, as shown in fig. 4, there is provided a wavelength compensation device including: an acquisition module 500, a prediction module 502, and a compensation module 504, wherein:
the obtaining module 500 is configured to obtain a tristimulus value corresponding to a wavelength to be compensated in a display screen.
The prediction module 502 is configured to input the tristimulus values into the wavelength prediction model, and perform conversion processing from the tristimulus values to the wavelengths on the input tristimulus values through the wavelength prediction model to obtain corresponding predicted wavelengths; the wavelength prediction model is obtained by training according to the tri-stimulus values of the multiple samples and the known wavelengths of the samples corresponding to the tri-stimulus values of the multiple samples.
And the compensation module 504 is configured to perform wavelength compensation on the wavelength to be compensated according to the predicted wavelength.
In one embodiment, the above apparatus further comprises: the training module is used for acquiring the tristimulus values of the samples of the preset number acquired according to the display screen under various conditions and the known wavelengths of the samples of the preset number; the tristimulus values of the samples in the preset number correspond to the known wavelengths of the samples in the preset number one by one; normalizing the preset number of sample tristimulus values to obtain the preset number of normalized sample tristimulus values; according to the normalized sample tristimulus values, acquiring corresponding sample predicted wavelengths through a wavelength prediction model to be trained; and updating the wavelength prediction model to be trained according to the sample predicted wavelength and the error value corresponding to the known wavelength of the sample until the error value is within a preset numerical range to obtain the wavelength prediction model.
In an embodiment, the training module is specifically configured to obtain a preset number of measurement tristimulus values, which are sent by the light collection device and are collected by the display screen under multiple conditions; for each measured tristimulus value, converting the measured tristimulus value into corresponding red primary color stimulus value, green primary color stimulus value and blue primary color stimulus value to obtain a sample tristimulus value; and obtaining the tristimulus values of the samples with preset quantity according to the tristimulus values of the samples.
In an embodiment, the training module is specifically configured to normalize, for each sample tristimulus value, a red primary color stimulus value, a green primary color stimulus value, and a blue primary color stimulus value in the sample tristimulus value, respectively, to obtain normalized sample tristimulus values; the numerical values of the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the normalized sample tristimulus values are all larger than 0 and less than or equal to 1; and obtaining a preset number of normalized sample tristimulus values according to the plurality of normalized sample tristimulus values.
In an embodiment, the training module is specifically configured to, for each normalized sample tristimulus value, obtain an output result corresponding to the hidden layer according to a red primary color stimulus value, a green primary color stimulus value, and a blue primary color stimulus value in the normalized sample tristimulus value, and a first weight of a BP neural network prediction model to be trained; the first weight represents the corresponding weight from the input layer to the hidden layer; acquiring a sample prediction wavelength corresponding to the normalized sample tristimulus value according to the output result and the second weight of the BP neural network prediction model to be trained; the second weight represents the weight corresponding to the hidden layer to the output layer.
In an embodiment, the training module is specifically configured to obtain an error value between a predicted wavelength of the sample and a known wavelength of the sample corresponding to the normalized tristimulus value; judging whether the error value is less than or equal to a preset value; if not, updating the first weight and the second weight according to the error value and preset learning efficiency, returning to the step of obtaining an output result corresponding to the hidden layer according to the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the normalized sample tristimulus values and the first weight of the BP neural network prediction model to be trained, and executing next model training until the number of model training times reaches the preset number of times; if so, finishing training the BP neural network prediction model to be trained to obtain a wavelength prediction model.
In an embodiment, the training module is specifically configured to obtain a first product of a red primary color stimulus value and a first weight, a second product of a green primary color stimulus value and the first weight, and a third product of a blue primary color stimulus value and the first weight in the normalized sample tristimulus values; and acquiring an output result corresponding to the hidden layer according to the first product, the second product and the third product.
In an embodiment, the training module is specifically configured to obtain a first adjustment value according to a partial derivative of the error value to the first weight and a preset learning efficiency; acquiring a second adjustment value according to the partial derivative of the error value to the second weight and the preset learning efficiency; acquiring the sum of the first weight and the first adjusting value to obtain an updated first weight; and acquiring the sum of the second weight and the second adjusting value to obtain the updated second weight.
For specific definition of the wavelength compensation device, reference may be made to the above definition of the wavelength compensation method, which is not described herein again. The modules in the wavelength compensation device can be implemented in whole or in part by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 5. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing data such as wavelength and tristimulus values. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a wavelength compensation method.
Those skilled in the art will appreciate that the architecture shown in fig. 5 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In an embodiment, a computer device is provided, comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the wavelength compensation method described above when executing the computer program.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which computer program, when being executed by a processor, is adapted to carry out the above-mentioned wavelength compensation method.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A method of wavelength compensation, the method comprising:
acquiring a tristimulus value corresponding to a wavelength to be compensated in a display screen;
inputting the tristimulus values into a wavelength prediction model, and performing conversion processing from the tristimulus values to the wavelengths on the input tristimulus values through the wavelength prediction model to obtain corresponding predicted wavelengths; the wavelength prediction model is obtained by training a wavelength prediction model to be trained according to the multiple sample tristimulus values and the known sample wavelengths corresponding to the sample tristimulus values; the wavelength prediction model to be trained comprises an input layer, a hidden layer and an output layer; further comprising: training the wavelength prediction model to be trained, wherein the training process comprises the following steps: obtaining a first product of a red primary color stimulus value and a first weight, a second product of a green primary color stimulus value and the first weight, and a third product of a blue primary color stimulus value and the first weight in normalized sample tristimulus values; acquiring an output result corresponding to the hidden layer according to the first product, the second product and the third product; the output result is used for inputting the output layer to obtain a sample predicted wavelength output by the output layer, and when an error value between the sample predicted wavelength and the known wavelength of the sample is within a preset numerical range, the wavelength prediction model is obtained; the first weight represents the corresponding weight from the input layer to the hidden layer;
and according to the predicted wavelength, performing wavelength compensation on the wavelength to be compensated.
2. The method according to claim 1, wherein before inputting the tristimulus values to be compensated into the wavelength prediction model, the method further comprises:
acquiring the tristimulus values of a preset number of samples acquired according to a display screen under various conditions and the known wavelengths of the preset number of samples; the tristimulus values of the samples in the preset number correspond to the known wavelengths of the samples in the preset number one by one;
normalizing the preset number of sample tristimulus values to obtain a preset number of normalized sample tristimulus values;
according to the normalized sample tristimulus values, acquiring corresponding sample predicted wavelengths through a wavelength prediction model to be trained;
and updating the wavelength prediction model to be trained according to the sample predicted wavelength and the error value corresponding to the known wavelength of the sample until the error value is within a preset value range, so as to obtain the wavelength prediction model.
3. The method of claim 2, wherein the obtaining a preset number of sample tristimulus values collected from a display screen under a plurality of conditions comprises:
acquiring the preset number of measurement tristimulus values which are sent by the light ray acquisition equipment and acquired by the display screen under the various conditions;
for each measured tristimulus value, converting the measured tristimulus value into corresponding red primary color stimulus value, green primary color stimulus value and blue primary color stimulus value to obtain the sample tristimulus value;
and obtaining the sample tristimulus values of the preset number according to the plurality of sample tristimulus values.
4. The method of claim 3, wherein normalizing the predetermined number of sample tristimulus values to obtain a predetermined number of normalized sample tristimulus values comprises:
respectively normalizing the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the sample tristimulus values aiming at each sample tristimulus value to obtain normalized sample tristimulus values; the numerical values of the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the normalized sample tristimulus values are all larger than 0 and less than or equal to 1;
and obtaining the normalized sample tristimulus values of the preset number according to the plurality of normalized sample tristimulus values.
5. The method according to claim 3, wherein the wavelength prediction model to be trained is a BP neural network prediction model to be trained, and the BP neural network prediction model to be trained comprises an input layer, a hidden layer and an output layer;
the obtaining of the corresponding sample prediction wavelength through the wavelength prediction model to be trained according to the normalized sample tristimulus value comprises the following steps:
aiming at each normalized sample tristimulus value, acquiring an output result corresponding to the hidden layer according to a red primary color stimulus value, a green primary color stimulus value and a blue primary color stimulus value in the normalized sample tristimulus values and a first weight of a BP neural network prediction model to be trained;
obtaining a sample prediction wavelength corresponding to the normalized sample tristimulus value according to the output result and a second weight of the BP neural network prediction model to be trained; the second weight represents the weight from the hidden layer to the output layer;
the step of updating the wavelength prediction model to be trained according to the error value corresponding to the sample predicted wavelength and the known wavelength of the sample until the error value is within a preset value range to obtain the wavelength prediction model comprises the following steps:
obtaining an error value between the sample predicted wavelength and a sample known wavelength corresponding to the normalized sample tristimulus value;
judging whether the error value is smaller than or equal to a preset value;
if not, updating the first weight and the second weight according to the error value and preset learning efficiency, returning to the step of obtaining an output result corresponding to the hidden layer according to the red primary color stimulus value, the green primary color stimulus value and the blue primary color stimulus value in the normalized sample tristimulus values and the first weight of the BP neural network prediction model to be trained, and executing next model training until the number of model training times reaches preset times;
if so, finishing training the BP neural network prediction model to be trained to obtain the wavelength prediction model.
6. The method of claim 5, wherein updating the first weight and the second weight according to the error value and a predetermined learning efficiency comprises:
acquiring a first adjusting value according to the partial derivative of the error value to the first weight and the preset learning efficiency;
acquiring a second adjusting value according to the partial derivative of the error value to the second weight and the preset learning efficiency;
acquiring the sum of the first weight and the first adjusting value to obtain an updated first weight;
and acquiring the sum of the second weight and the second adjusting value to obtain the updated second weight.
7. A wavelength compensation device, the device comprising:
the acquisition module is used for acquiring the tristimulus values corresponding to the wavelengths to be compensated in the display screen;
the prediction module is used for inputting the tristimulus values into a wavelength prediction model, and performing conversion processing from the tristimulus values to the wavelengths on the input tristimulus values through the wavelength prediction model to obtain corresponding predicted wavelengths; the wavelength prediction model is obtained by training a wavelength prediction model to be trained according to the multiple sample tristimulus values and the known sample wavelengths corresponding to the sample tristimulus values; the wavelength prediction model to be trained comprises an input layer, a hidden layer and an output layer; further comprising: a training module, configured to train the wavelength prediction model to be trained, where the training module is specifically configured to: obtaining a first product of a red primary color stimulus value and a first weight, a second product of a green primary color stimulus value and the first weight, and a third product of a blue primary color stimulus value and the first weight in normalized sample tristimulus values; acquiring an output result corresponding to the hidden layer according to the first product, the second product and the third product; the output result is used for inputting the output layer to obtain a sample predicted wavelength output by the output layer, and when an error value between the sample predicted wavelength and the known wavelength of the sample is within a preset numerical range, the wavelength prediction model is obtained; the first weight represents the corresponding weight from the input layer to the hidden layer;
and the compensation module is used for carrying out wavelength compensation on the wavelength to be compensated according to the predicted wavelength.
8. The apparatus of claim 7, further comprising:
the training module is used for acquiring the tristimulus values of the samples of the preset number acquired according to the display screen under various conditions and the known wavelengths of the samples of the preset number; the tristimulus values of the samples in the preset number correspond to the known wavelengths of the samples in the preset number one by one; normalizing the preset number of sample tristimulus values to obtain the preset number of normalized sample tristimulus values; according to the normalized sample tristimulus values, acquiring corresponding sample predicted wavelengths through a wavelength prediction model to be trained; and updating the wavelength prediction model to be trained according to the sample predicted wavelength and the error value corresponding to the known wavelength of the sample until the error value is within a preset numerical range to obtain the wavelength prediction model.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110288386.0A CN112801216B (en) | 2021-03-18 | 2021-03-18 | Wavelength compensation method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110288386.0A CN112801216B (en) | 2021-03-18 | 2021-03-18 | Wavelength compensation method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112801216A CN112801216A (en) | 2021-05-14 |
| CN112801216B true CN112801216B (en) | 2021-07-20 |
Family
ID=75817139
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110288386.0A Active CN112801216B (en) | 2021-03-18 | 2021-03-18 | Wavelength compensation method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112801216B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114544629B (en) * | 2022-02-22 | 2024-05-24 | 西安交通大学 | Method for judging specific number of layers of graphene within ten layers according to optical microscope image |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104215333B (en) * | 2013-06-04 | 2016-08-10 | 致茂电子(苏州)有限公司 | Two dimension formula sequential type colorimeter detection method and this colorimeter |
| CN107113410A (en) * | 2015-01-19 | 2017-08-29 | 三菱电机株式会社 | Video display devices and display bearing calibration |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101527108B (en) * | 2009-04-17 | 2012-12-12 | 天津光电星球显示设备有限公司 | Measuring method for colorimetric parameters of LED display screen |
| JP7207124B2 (en) * | 2019-04-16 | 2023-01-18 | セイコーエプソン株式会社 | CALIBRATION DEVICE, CALIBRATION METHOD, SPECTRAL CAMERA, AND DISPLAY DEVICE |
-
2021
- 2021-03-18 CN CN202110288386.0A patent/CN112801216B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104215333B (en) * | 2013-06-04 | 2016-08-10 | 致茂电子(苏州)有限公司 | Two dimension formula sequential type colorimeter detection method and this colorimeter |
| CN107113410A (en) * | 2015-01-19 | 2017-08-29 | 三菱电机株式会社 | Video display devices and display bearing calibration |
Non-Patent Citations (1)
| Title |
|---|
| 颜色测量及其在光谱测量中的应用研究;张卉;《中国优秀硕士学位论文全文数据库基础科学辑》;20120215(第02期);第A005-36页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112801216A (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2161555B1 (en) | Coating color database creating method, search method using the database, their system, program, and recording medium | |
| US10568178B2 (en) | Color temperature adjustment method and device, light source, display apparatus | |
| CN110555465B (en) | Weather image identification method based on CNN and multi-feature fusion | |
| CN113284456B (en) | Method and device for determining correction coefficient of display screen lamp point and computer equipment | |
| CN104320641B (en) | Apparatus and method for automated self-training of white balance by electronic cameras | |
| CN110675818B (en) | Curve matching-based module Gamma correction method and system | |
| CN115547270A (en) | Chromatic aberration adjusting method, device and equipment based on spectral analysis and storage medium | |
| CN112840636A (en) | Image processing method and device | |
| CN112801216B (en) | Wavelength compensation method and device, computer equipment and storage medium | |
| EP3888345A1 (en) | Method for generating image data for machine learning based imaging algorithms | |
| CN110310596A (en) | A kind of the GAMMA adjusting initial value prediction technique and system of OLED mould group | |
| CN104092919B (en) | Chromatic adaptation transformation optimizing method and system for color digital imaging system | |
| CN114512085A (en) | Visual color calibration method of TFT (thin film transistor) display screen | |
| CN112967669A (en) | Correction method for improving gray scale consistency of LED display screen | |
| CN109035223A (en) | A kind of intelligent evaluation method for satellite remote sensing images availability | |
| CN117320219A (en) | Fluorescent powder-free multi-primary-color LED lamp illumination control method and system | |
| CN109308874A (en) | Display screen brightness adjusting method and device | |
| CN117975876A (en) | Display screen correction method, device, computer equipment and storage medium | |
| US20230056947A1 (en) | Learning apparatus, method, and storage medium | |
| CN113487577A (en) | GRU-CNN combined model-based rapid Gamma adjustment method, system and application | |
| CN106408433A (en) | Grain yield prediction method and grain yield prediction device based on differential processing | |
| CN105631812A (en) | Control method and control device for performing color enhancement on displayed image | |
| Talun et al. | Test Images for Training Convolutional Neural Networks for Image Contrast Assessment | |
| CN117768791B (en) | Cross-camera image white balance method and system based on neural network | |
| CN116188797A (en) | Scene light source color estimation method capable of being effectively embedded into image signal processor |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |