CN118037602A - Image quality optimization method, device, electronic equipment, medium and program product - Google Patents
Image quality optimization method, device, electronic equipment, medium and program product Download PDFInfo
- Publication number
- CN118037602A CN118037602A CN202410448087.2A CN202410448087A CN118037602A CN 118037602 A CN118037602 A CN 118037602A CN 202410448087 A CN202410448087 A CN 202410448087A CN 118037602 A CN118037602 A CN 118037602A
- Authority
- CN
- China
- Prior art keywords
- image
- evaluation
- image quality
- evaluated
- optimization
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The application discloses an image quality optimization method, an image quality optimization device, electronic equipment, a medium and a program product, and relates to the technical field of image processing, wherein the image quality optimization method comprises the following steps: obtaining a standard image corresponding to an image to be evaluated, and determining a target image quality evaluation model corresponding to the standard image; inputting the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, wherein the evaluation result comprises one of qualification and optimization; when the evaluation result is to be optimized, generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated, and optimizing the image to be evaluated based on the image optimization strategy. The application improves the optimization efficiency of the image quality.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to an image quality optimization method, an image quality optimization device, an electronic device, a computer readable storage medium, and a computer program product.
Background
With the development of camera technology, image shooting has become one of the functions commonly used by a large number of users, and users can shoot landscapes and people through mobile phones, cameras or other devices with shooting functions. However, due to reasons of ambient light of photographing or improper setting of photographing parameters, the quality of the finally obtained image may not be satisfactory, and further adjustment and optimization are required. The current commonly used image optimization method is that a user optimizes the image quality by adjusting image parameters by using picture processing software. However, the optimization mode is separated from the shooting step, the user is required to additionally perform an image quality optimization step after obtaining the image, the image optimization processing result depends on the control and adjustment capability of the user on the image parameters, the threshold is high, the time consumption is long, and therefore the image optimization efficiency is low.
Disclosure of Invention
The main object of the present application is to provide an image quality optimization method, apparatus, electronic device, computer readable storage medium and computer program product, aiming at improving the optimization efficiency of image quality.
To achieve the above object, the present application provides an image quality optimization method comprising:
Obtaining a standard image corresponding to an image to be evaluated, and determining a target image quality evaluation model corresponding to the standard image;
Inputting the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, wherein the evaluation result comprises one of qualification and optimization;
When the evaluation result is to be optimized, generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated, and optimizing the image to be evaluated based on the image optimization strategy.
Optionally, the image optimization strategy is an image processing strategy, and the step of generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated includes:
inquiring an image processing strategy corresponding to the defect type in a preset image processing strategy library;
And adjusting the image parameters of the image to be evaluated based on the image processing strategy to generate a target optimized image.
Optionally, before the step of inputting the image to be evaluated into the target image quality evaluation model, the method further comprises:
Acquiring evaluation labels corresponding to a plurality of training images respectively, and setting each evaluation label and the corresponding training image as an evaluation subtask, wherein the evaluation labels comprise subjective evaluation results corresponding to each training image by a plurality of evaluation staff respectively;
And carrying out iterative training on a preset initial image quality assessment model based on each assessment subtask to obtain a target image quality assessment model.
Optionally, the step of performing iterative training on a preset initial image quality assessment model based on each assessment subtask to obtain a target image quality assessment model includes:
acquiring gradients of all the evaluation subtasks;
Screening a plurality of evaluation tasks based on the gradient of each evaluation subtask to obtain a training set, wherein the evaluation tasks comprise a plurality of evaluation subtasks corresponding to each evaluation person;
sequentially selecting an evaluation task from the training set according to a preset sequence to perform iterative training on the initial image quality evaluation model, and acquiring a loss function value of the initial image quality evaluation model in the iterative training process;
and stopping training when the loss function value is converged, and setting the current initial image quality assessment model as a target image quality assessment model.
Optionally, the step of screening the plurality of evaluation tasks based on the gradient of each evaluation subtask to obtain a training set includes:
According to gradients of a plurality of evaluation subtasks corresponding to each evaluation personnel, a gradient matrix of the evaluation tasks corresponding to each evaluation personnel is obtained;
determining a similarity value between the gradient matrix of each evaluator and the gradient matrix of other evaluators in the evaluator set;
And screening a preset number of evaluation tasks with front similarity values from the evaluation tasks to form a training set.
Optionally, the evaluation result further includes non-optimization, and after the step of obtaining the evaluation result and/or defect type corresponding to the image to be evaluated, the method further includes:
If the evaluation result of the image to be evaluated is that optimization is impossible, inquiring a shooting parameter adjustment strategy corresponding to the defect type in a preset shooting parameter adjustment strategy library based on the defect type corresponding to the image to be evaluated;
And carrying out parameter optimization on initial shooting parameters of shooting equipment corresponding to the image to be evaluated based on the shooting parameter adjustment strategy.
The present application also provides an image quality optimizing apparatus including:
The model determining module is used for acquiring a standard image corresponding to the image to be evaluated and determining a target image quality evaluation model corresponding to the standard image;
The image evaluation module is used for inputting the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, wherein the evaluation result comprises one of qualification and optimization;
And the image optimization module is used for generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated when the evaluation result is to be optimized, and optimizing the image to be evaluated based on the image optimization strategy.
The application also provides an electronic device, which is entity equipment, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the image quality optimization method as described above.
The present application also provides a readable storage medium which is a computer readable storage medium having stored thereon a program for implementing an image quality optimization method, the program for implementing the image quality optimization method being executed by a processor to implement the steps of the image quality optimization method as described above.
The application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the image quality optimization method as described above.
The application provides an image quality optimization method, which comprises the steps of firstly obtaining a standard image corresponding to an image to be evaluated, determining a target image quality evaluation model corresponding to the standard image, subdividing the image quality evaluation models under the condition of different standard images, improving the prediction precision of the image quality evaluation models, inputting the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, wherein the evaluation result comprises one of qualification and optimization to be performed, realizing automatic evaluation of the image through a pre-trained deep learning model, immediately outputting the corresponding evaluation result at the first time of generation of the image to be evaluated, generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated when the evaluation result is the optimization to be performed, and optimizing the image to be evaluated based on the image optimization strategy. According to the technical scheme, when the image to be evaluated needs to be optimized, a corresponding image optimization strategy is automatically formulated according to the defect type, so that the automatic optimization of the image is realized, the technical defects of long time consumption, low efficiency and the like caused by manually performing image quality optimization are overcome, and the optimization efficiency and the optimization effect on the image quality are improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the description of the embodiments or the prior art will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.
Fig. 1 is a flowchart of an image quality optimization method according to a first embodiment of the present application;
fig. 2 is a schematic flow chart of steps a10 to a20 in the second embodiment of the present application;
fig. 3 is a schematic flow chart of steps a21 to a24 in the second embodiment of the present application;
FIG. 4 is a schematic diagram of an image quality optimization apparatus according to an embodiment of the present application;
Fig. 5 is a schematic device structure diagram of a hardware operating environment related to an image quality optimization method according to an embodiment of the present application.
The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
In order to make the above objects, features and advantages of the present application more comprehensible, the following description of the embodiments accompanied with the accompanying drawings will be given in detail. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Example 1
An embodiment of the present application provides an image quality optimization method, please refer to fig. 1, which includes the steps of:
Step S10, a standard image corresponding to an image to be evaluated is obtained, and a target image quality evaluation model corresponding to the standard image is determined;
it should be noted that the image quality optimization method may be applied to a camera, a mobile phone, or other devices having a photographing function. The standard image is a preset high-quality image used for being used as a reference in different scenes. It will be appreciated that in different scenarios, the style of the images that the user needs to capture is inconsistent, e.g., sunlight, silence, macro or sweet, etc. In each scene, one or more standard images preset by a user are provided, and the standard images represent the excellent image parameter adjustment and composition in the current scene and the current style. In addition, the target image quality evaluation model corresponding to the standard image is a pre-trained deep learning network model, which is used for evaluating whether the image to be evaluated needs to be optimized and has defects based on the standard image. Therefore, with multiple selectable standard images, there are also the same number of corresponding target image quality assessment models.
For example, when obtaining a standard image corresponding to an image to be evaluated, a user may select one standard image corresponding to a direction and style that are required to be optimized for the current image to be evaluated. Compared with the method that a user determines a target image quality evaluation model through the style of the word description, the method has the advantages that the user can more intuitively browse the image effect by selecting the standard image in the image form, clearly know the required optimized image style direction, and avoid low satisfaction degree on the finally obtained optimized image caused by misunderstanding or unclear understanding of the word description.
Step S20, inputting the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, wherein the evaluation result comprises one of qualification and optimization;
In the embodiment of the application, after the corresponding target image quality evaluation model is determined through the standard image, the image to be evaluated can be input into the target image quality evaluation model, so that the quality evaluation of the image to be evaluated is performed through the target image quality evaluation model, and the corresponding evaluation result is output. The evaluation result can be qualified or can be optimized, the qualification means that the difference between the image to be evaluated and the standard image is smaller, the expected difference accords with the expectation, the optimization is not needed, the difference between the image to be evaluated and the standard image is larger, the expected difference does not accord with the expectation, and the further optimization adjustment is needed. Further, when the evaluation result is to be optimized, the target image quality evaluation model synchronously outputs the defect types of the image to be evaluated compared with the standard image, such as too low brightness, low definition, low saturation, cold-warm degree deviation, contrast deviation, composition main body position deviation and the like. The defect type of the image to be evaluated is determined, so that a reliable basis can be provided for further image optimization, and the pertinence and the optimization effect of the image optimization are improved.
It should be noted that the target image evaluation model is obtained by performing iterative optimization training on a large number of training images, subjective evaluation and labeling can be performed on the training images by multiple evaluation staff in advance, and particularly, based on the difference between the training images and the standard images, the training images are labeled with labels to be qualified or optimized, and the corresponding defect types are labeled while the labels to be optimized are labeled. Therefore, the target image evaluation model can learn the evaluation standard of an evaluation person on the image in the training process, and finally, the image to be evaluated can be accurately evaluated and the defect detection can be carried out.
And step S30, when the evaluation result is to be optimized, generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated, and optimizing the image to be evaluated based on the image optimization strategy.
It can be understood that when the output evaluation result is to be optimized, the corresponding image optimization policy may be determined according to the defect type corresponding to the image to be evaluated, and specifically, the image optimization policy (for example, the image parameter adjustment policy, etc.) corresponding to the defect type may be queried in the preset two-dimensional mapping table.
In addition, in another possible embodiment, when the evaluation result of the image to be evaluated is output as the image to be optimized, the target image quality evaluation model may output specific defect information, such as a luminance of 20% or a saturation of 15% or a subject of 30% or more in the composition, in addition to the corresponding defect type. More specific defect information can provide more accurate data reference for generating an image optimization strategy, obtain the image optimization strategy with higher accuracy, and output an optimized image with better image quality.
The embodiment of the application provides an image quality optimization method, which comprises the steps of firstly obtaining a standard image corresponding to an image to be evaluated, determining a target image quality evaluation model corresponding to the standard image, subdividing the image quality evaluation models under the condition of different standard images, improving the prediction precision of the image quality evaluation models, inputting the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, wherein the evaluation result comprises one of qualification and optimization to be performed, automatic evaluation of the image through a pre-trained deep learning model is realized, the corresponding evaluation result can be output immediately at the first time generated by the image to be evaluated, when the evaluation result is the optimization to be performed, a corresponding image optimization strategy is generated according to the defect type corresponding to the image to be evaluated, and the image to be evaluated is optimized based on the image optimization strategy. According to the technical scheme provided by the embodiment of the application, when the image to be evaluated needs to be optimized, a corresponding image optimization strategy is automatically formulated according to the defect type, so that the automatic optimization of the image is realized, the technical defects of long time consumption, low efficiency and the like caused by manually performing image quality optimization are overcome, and the optimization efficiency and the optimization effect on the image quality are improved.
Further, in a possible embodiment, the image optimization policy is an image processing policy, and the step of generating a corresponding image optimization policy according to the defect type corresponding to the image to be evaluated may include:
step S31, inquiring an image processing strategy corresponding to the defect type in a preset image processing strategy library;
According to the technical scheme, when the output evaluation result is to be optimized to be the defect type, further image processing and optimization can be performed on the image to be evaluated, firstly, the corresponding image processing strategies can be queried according to the defect type in a preset image processing strategy library, wherein the image processing strategy library can comprise a two-dimensional mapping table, the two-dimensional mapping table represents the mapping relation between various defect types and the corresponding image processing strategies, the proper image processing strategies can be determined according to each defect type, and the image processing strategies are parameter adjustment strategies, and can comprise adjustment directions, adjustment amplitudes and the like of various parameters. The two-dimensional mapping table is produced and generated by a worker according to actual image adjustment optimization experience in advance.
And step S32, adjusting the image parameters of the image to be evaluated based on the image processing strategy to generate a target optimized image.
In the embodiment of the application, the mobile phone, the camera and other devices applying the image quality optimization method are internally provided with a processor and a memory, the processor can call an image processing algorithm preset in the memory and carry out parameter adjustment on the image parameters of the image to be evaluated based on the determined image processing strategy until a target optimization image is generated and output, wherein the target optimization image is the image to be evaluated after parameter adjustment is completed.
In a possible embodiment, in the process of adjusting parameters of the image to be evaluated, the target image quality evaluation model may be called again to evaluate the result of parameter adjustment, and if the evaluation result after parameter adjustment is still to be optimized, the adjustment is continued until the output evaluation result is qualified.
The technical scheme of the embodiment of the application mainly queries the corresponding image processing strategy according to the defect type output by the target image quality evaluation model, realizes the adoption of a corresponding accurate processing mode aiming at different defect types, overcomes the technical defects of overhigh threshold and excessively long time consumption of manually performing image parameter adjustment, and improves the image quality optimization efficiency.
Further, in a possible embodiment, the evaluation result further includes non-optimization, and after the step of obtaining the evaluation result and/or the defect type corresponding to the image to be evaluated, the method may further include:
step S40, if the evaluation result of the image to be evaluated is that optimization is impossible, inquiring a shooting parameter adjustment strategy corresponding to the defect type in a preset shooting parameter adjustment strategy library based on the defect type corresponding to the image to be evaluated;
And step S50, carrying out parameter optimization on initial shooting parameters of shooting equipment corresponding to the image to be evaluated based on the shooting parameter adjustment strategy.
It should be noted that, in some cases, the image to be evaluated cannot obtain an optimized image meeting the expected target by means of image parameter adjustment, which may be caused by improper parameter adjustment when the photographing device performs photographing, for example, the brightness setting is too low, enough light is not collected to generate a bright image, and the distortion and the sharpness are reduced by directly adjusting the brightness of the image. On the other hand, there is also a possibility of image blurring due to a problem of improper angle or shake at the time of photographing, which cannot be done optimally by adjusting the image parameters of the image to be evaluated. Therefore, by adjusting the shooting parameters, the technical scheme of the embodiment of the application enables the user to carry out shooting under the relatively proper shooting parameters in the next shooting process, and obtains the image with higher image quality.
Specifically, similar to the step S31, the step S40 is also obtained by searching in a two-dimensional mapping table of a preset shooting parameter adjustment policy library when determining the shooting parameter adjustment policy, which is not described herein. In addition, after the shooting parameter adjustment strategy is obtained, before the initial shooting parameters of the shooting equipment are adjusted based on the shooting parameter adjustment strategy, prompt information can be sent to a user through a display interface so as to prompt the user that the shooting parameters are currently and automatically adjusted, and the user can confirm the initial shooting parameters through a pop-up dialog box.
In another possible embodiment, when the defect type of the image to be evaluated is determined to be image blurring caused by improper angles or shaking, the image blurring cannot be optimized by adjusting shooting parameters. At this time, the optimization strategy corresponding to the defect type can be output through text or voice prompt and other modes. For example: the camera lens is adjusted, the shooting main body is arranged at the central position of the picture, the shooting equipment is stabilized, and the like, so that a user can timely determine and correct the shooting problems existing in the camera lens, and the quality of the shot image is improved.
The embodiment of the application provides an image optimization method when an image to be evaluated cannot be optimized by adjusting image parameters, namely, parameter adjustment is carried out on shooting equipment by determining a shooting parameter adjustment strategy, so that the image quality of the next shooting is improved. In addition, under the condition that the image quality can not be optimized by adjusting the parameters of the shooting equipment, prompt information is output in real time, so that a user can correct shooting actions in time. The automatic optimization of the quality of the photographed image is realized, the photographing threshold of the user is further reduced, and even if the user does not have enough photographing experience, the target image with relatively high image quality can be obtained.
Example two
In another embodiment of the present application, the same or similar content as that of the first embodiment can be referred to the description above, and the description is omitted. On this basis, before the step of inputting the image to be evaluated into the target image quality evaluation model, referring to fig. 2, the method may further include:
Step A10, acquiring evaluation labels corresponding to a plurality of training images respectively, and setting each evaluation label and the corresponding training image as an evaluation subtask, wherein the evaluation labels comprise subjective evaluation results corresponding to the training images respectively by a plurality of evaluation staff;
and step A20, performing iterative training on a preset initial image quality assessment model based on each assessment subtask to obtain a target image quality assessment model.
The embodiment of the application provides a method for training a target image quality evaluation model, which comprises the steps of firstly evaluating and labeling a plurality of training images by a plurality of evaluation personnel, wherein in the process of evaluating the training images, a standard image is used for reference, and an evaluation result (qualified or to be optimized) is obtained by judging the difference between the training images and the standard image, the obtained evaluation label comprises the evaluation result and/or defect type, and the target image quality evaluation model obtained by corresponding to training is also corresponding to the standard image and is only used for selecting the image quality evaluation under the standard image by a user.
Specifically, the evaluation labels of the images are results obtained by subjective evaluation of K images by N evaluators, wherein K and N are positive integers greater than or equal to 1. The K images can be selected from images under different scenes, such as images with different quality, such as brightness, angle, exposure degree and the like, acquired in image shooting, the N evaluators evaluate the K images with different quality respectively, and each training image and a corresponding evaluation label serve as a separate evaluation subtask. It should be noted that, each training image is evaluated by multiple persons, so each training image may correspond to multiple evaluation subtasks.
As an example, the preset initial image quality assessment model may be a meta-learning model. The meta learning model is a machine learning model, is a model for solving the problems of insufficient generalization performance and poor adaptability to different kinds of tasks existing in a common neural network model, can quickly learn a new concept through a small number of data samples or can be well adapted and generalized to a new task after training of different tasks. Therefore, in the embodiment of the application, because factors influencing the image quality are more when the image is shot, the images in all cases cannot be enumerated by using the general neural network model, so that the general neural network model is difficult to evaluate the image quality.
Further, in a possible embodiment, referring to fig. 3, the step of performing iterative training on the preset initial image quality assessment model based on each of the assessment subtasks to obtain the target image quality assessment model may include:
Step A21, obtaining the gradient of each evaluation subtask;
It should be noted that the gradient is a vector, and the partial derivative used to find the optimal parameter in the meta-learning model, that is, in the meta-learning model, each evaluation subtask may be expressed by means of a gradient.
Step A22, screening a plurality of evaluation tasks based on the gradient of each evaluation subtask to obtain a training set, wherein the evaluation tasks comprise a plurality of evaluation subtasks corresponding to each evaluation personnel;
when screening the evaluation tasks based on the gradients of the evaluation subtasks, firstly, acquiring the gradient of each evaluation subtask, and determining the gradient combination of K evaluation subtasks corresponding to each evaluation person based on the gradient of each evaluation subtask to obtain a gradient matrix of each evaluation person evaluation task, wherein the gradient matrix corresponds to the evaluation tasks one by one. And screening the evaluation tasks corresponding to the gradient matrixes with higher similarity values according to the similarity value between each gradient matrix and all other gradient matrixes to form a training set.
Step A23, sequentially selecting an evaluation task from the training set to perform iterative training on the initial image quality evaluation model according to a preset sequence, and acquiring a loss function value of the initial image quality evaluation model in the iterative training process;
The preset sequence is the sequence from high to low in similarity value, and the higher the similarity value is, the stronger the representativeness of the evaluation result corresponding to the evaluator with respect to all the evaluators is, and the higher the evaluation task value corresponding to the evaluation result is. Therefore, the technical scheme of the embodiment of the application preferentially selects the evaluation task with higher value to perform model training, so that the model can learn the standard of image quality evaluation by an evaluator faster, and a target image quality evaluation model with higher prediction precision, more stable performance and stronger generalization capability is obtained. In the training process, the loss function value is calculated mainly based on the prediction result and the evaluation label output by the initial image quality evaluation model, and the model parameters of the initial image quality evaluation model are optimized and adjusted so that the loss function value changes to the direction of reducing as much as possible.
It will be appreciated that when the loss function value does not converge, the loop continues to perform step a23 until the loss function value converges.
And step A24, stopping training when the loss function value is converged, and setting the current initial image quality assessment model as a target image quality assessment model.
The loss function is used for measuring the inconsistency degree of the predicted value and the true value of the model, the loss function value is smaller along with the iterative training of the model, the predicted value representing the model is closer to the true value, and when the loss function value of the model is not reduced any more, the loss function can be considered to be converged.
In the above embodiment, the evaluation tasks are screened by using the gradient of the evaluation subtasks, and the screened evaluation tasks are used as the training set of the meta-learning model, so that the training set is more accurately selected, the evaluation tasks are selected from the training set according to the preset sequence, and the meta-learning model is iteratively trained until the loss function of the meta-learning model is converged, so that the evaluation model with stable performance and prediction precision meeting the expected target image quality can be obtained.
Further, in a possible embodiment, the step of screening the plurality of evaluation tasks to obtain the training set based on the gradient of each of the evaluation subtasks may include:
Step A221, obtaining a gradient matrix of the evaluation task corresponding to each evaluation person according to the gradients of the plurality of evaluation subtasks corresponding to each evaluation person;
Each evaluator corresponds to K evaluation subtasks, that is, each evaluator corresponds to the gradient of K evaluation subtasks, and K evaluation subtasks corresponding to one evaluator constitute the evaluation task of the evaluator, so that the gradient matrix of the evaluation task of the evaluator can be obtained by combining the gradients of the K evaluation subtasks of each evaluator.
Step A222, determining a similarity value between the gradient matrix of each evaluator and the gradient matrix of other evaluators in the evaluator set;
Specifically, similarity calculation is required to be performed on the gradient matrix of each evaluator in the N evaluators and the gradient matrix of the remaining evaluators in the N evaluators (i.e., the evaluator set) respectively, so as to obtain a similarity value between the evaluation task of each evaluator and the evaluation task of the remaining evaluators;
Illustratively, the similarity value of the a-th evaluator The calculation formula of (2) is as follows:
；
Wherein R is a set of all evaluation subtasks corresponding to all evaluation personnel, ai is an integer of more than or equal to 1 and less than or equal to N, Is the gradient matrix corresponding to the a-th evaluators,/>And/>The gradient matrix of the evaluation task of all the remaining evaluators is shown, and j and h are integers greater than or equal to 1 and less than or equal to K.
Specifically, the higher the similarity value of the gradient matrix of the evaluator is, the higher the similarity between the gradient matrix corresponding to the evaluator and the gradient matrix of the rest of evaluators is, so that the evaluator is more representative of the image quality evaluation result.
And step A223, selecting a preset number of evaluation tasks with front similarity values from the evaluation tasks to form a training set.
Firstly, sorting similarity values of evaluation tasks of all evaluation personnel, and screening M evaluation personnel corresponding to similarity values with the front sorting; wherein M is a positive integer of 1 to N.
Specifically, the similarity value w= { of N evaluators,/>,… ,/>Sorting W: wrank = sort (W), where sort is a descending order ordering algorithm. And the evaluators corresponding to the similarity values ranked at the front are evaluators with high similarity values. And finally, taking the evaluation tasks of M evaluation personnel as a training set of the meta learning model. Wherein, each evaluation personnel's of M evaluation personnel evaluation task has included K evaluation subtasks.
In the above embodiment, all evaluation subtasks corresponding to each evaluator, namely, K subtasks, use gradients of the K evaluation subtasks corresponding to each evaluator to form a gradient matrix of each human evaluation task, use the gradient matrix of the evaluation task of each evaluator to calculate and obtain similarity values between the evaluation task of each evaluator and the evaluation tasks of the rest evaluators, and then screen out M persons with high similarity values according to the sorting result of the similarity values, and use the evaluation tasks of the M persons as a training set, namely, screen out a representative evaluation task as a training set, thereby facilitating training of a meta-learning model and improving model training efficiency.
Example III
An embodiment of the present application provides an image quality optimization apparatus, referring to fig. 4, including:
the model determining module 10 is configured to obtain a standard image corresponding to an image to be evaluated, and determine a target image quality evaluation model corresponding to the standard image;
The image evaluation module 20 is configured to input the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, where the evaluation result includes one of qualification and optimization;
and the image optimization module 30 is configured to generate a corresponding image optimization policy according to a defect type corresponding to the image to be evaluated when the evaluation result is to be optimized, and optimize the image to be evaluated based on the image optimization policy.
Optionally, the image optimization policy is an image processing policy, and the image optimization module 30 is further configured to:
inquiring an image processing strategy corresponding to the defect type in a preset image processing strategy library;
And adjusting the image parameters of the image to be evaluated based on the image processing strategy to generate a target optimized image.
Optionally, the image quality optimization device further comprises a model training module, wherein the model training module is used for:
Acquiring evaluation labels corresponding to a plurality of training images respectively, and setting each evaluation label and the corresponding training image as an evaluation subtask, wherein the evaluation labels comprise subjective evaluation results corresponding to each training image by a plurality of evaluation staff respectively;
And carrying out iterative training on a preset initial image quality assessment model based on each assessment subtask to obtain a target image quality assessment model.
Optionally, the model training module is further configured to:
acquiring gradients of all the evaluation subtasks;
Screening a plurality of evaluation tasks based on the gradient of each evaluation subtask to obtain a training set, wherein the evaluation tasks comprise a plurality of evaluation subtasks corresponding to each evaluation person;
sequentially selecting an evaluation task from the training set according to a preset sequence to perform iterative training on the initial image quality evaluation model, and acquiring a loss function value of the initial image quality evaluation model in the iterative training process;
and stopping training when the loss function value is converged, and setting the current initial image quality assessment model as a target image quality assessment model.
Optionally, the model training module is further configured to:
According to gradients of a plurality of evaluation subtasks corresponding to each evaluation personnel, a gradient matrix of the evaluation tasks corresponding to each evaluation personnel is obtained;
determining a similarity value between the gradient matrix of each evaluator and the gradient matrix of other evaluators in the evaluator set;
And screening a preset number of evaluation tasks with front similarity values from the evaluation tasks to form a training set.
Optionally, the evaluation result further includes non-optimizations, and the image optimization module 30 is further configured to:
If the evaluation result of the image to be evaluated is that optimization is impossible, inquiring a shooting parameter adjustment strategy corresponding to the defect type in a preset shooting parameter adjustment strategy library based on the defect type corresponding to the image to be evaluated;
And carrying out parameter optimization on initial shooting parameters of shooting equipment corresponding to the image to be evaluated based on the shooting parameter adjustment strategy.
The image quality optimization device provided by the embodiment of the application can solve the technical problem of lower optimization efficiency of image quality by adopting the image quality optimization method in the embodiment. Compared with the prior art, the beneficial effects of the image quality optimization device provided by the embodiment of the application are the same as those of the image quality optimization method provided by the embodiment, and other technical features of the image quality optimization device are the same as those disclosed by the method of the embodiment, so that details are not repeated.
Example IV
The embodiment of the application provides electronic equipment, which comprises: at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can perform the image quality optimization method in the first embodiment.
Referring now to fig. 5, a schematic diagram of an electronic device suitable for use in implementing embodiments of the present disclosure is shown. The electronic devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal DIGITAL ASSISTANT: personal digital assistants), PADs (Portable Application Description: tablet computers), PMPs (Portable MEDIA PLAYER: portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 5 is merely an example and should not be construed to limit the functionality and scope of use of the disclosed embodiments.
As shown in fig. 5, the electronic apparatus may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access Memory (RAM: random Access Memory) 1004. In the RAM1004, various programs and data required for the operation of the electronic device are also stored. The processing device 1001, the ROM1002, and the RAM1004 are connected to each other by a bus 1005. An input/output (I/O) interface 1006 is also connected to the bus. In general, the following systems may be connected to the I/O interface 1006: input devices 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, and the like; an output device 1008 including, for example, a Liquid crystal display (LCD: liquid CRYSTAL DISPLAY), a speaker, a vibrator, and the like; storage device 1003 including, for example, a magnetic tape, a hard disk, and the like; and communication means 1009. The communication means 1009 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. While electronic devices having various systems are shown in the figures, it should be understood that not all of the illustrated systems are required to be implemented or provided. More or fewer systems may alternatively be implemented or provided.
In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network through a communication device, or installed from the storage device 1003, or installed from the ROM 1002. The above-described functions defined in the method of the embodiment of the present disclosure are performed when the computer program is executed by the processing device 1001.
The electronic equipment provided by the application can solve the technical problem of lower optimization efficiency of image quality by adopting the image quality optimization method in the embodiment. Compared with the prior art, the electronic device provided by the embodiment of the application has the same beneficial effects as the image quality optimization method provided by the embodiment, and other technical features in the electronic device are the same as the features disclosed by the method of the previous embodiment, and are not described in detail herein.
It should be understood that portions of the present disclosure may be implemented in hardware, software, firmware, or a combination thereof. In the description of the above embodiments, particular features, structures, materials, or characteristics may be combined in any suitable manner in any one or more embodiments or examples.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Example five
An embodiment of the present application provides a computer-readable storage medium having computer-readable program instructions stored thereon for performing the image quality optimization method of the first embodiment.
The computer readable storage medium according to the embodiments of the present application may be, for example, a usb disk, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access Memory (RAM: random Access Memory), a Read-Only Memory (ROM: read Only Memory), an erasable programmable Read-Only Memory (EPROM: erasable Programmable Read Only Memory or flash Memory), an optical fiber, a portable compact disc Read-Only Memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this embodiment, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, or device. Program code embodied on a computer readable storage medium may be transmitted using any appropriate medium, including but not limited to: wire, fiber optic cable, RF (Radio Frequency), and the like, or any suitable combination of the foregoing.
The above-described computer-readable storage medium may be contained in an electronic device; or may exist alone without being assembled into an electronic device.
The computer-readable storage medium carries one or more programs that, when executed by an electronic device, cause the electronic device to: obtaining a standard image corresponding to an image to be evaluated, and determining a target image quality evaluation model corresponding to the standard image; inputting the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, wherein the evaluation result comprises one of qualification and optimization; when the evaluation result is to be optimized, generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated, and optimizing the image to be evaluated based on the image optimization strategy.
Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of remote computers, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN: local Area Network) or a wide area network (WAN: wide Area Network), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules described in the embodiments of the present disclosure may be implemented in software or hardware. Wherein the name of the module does not constitute a limitation of the unit itself in some cases.
The readable storage medium provided by the application is a computer readable storage medium, and the computer readable storage medium stores computer readable program instructions for executing the image quality optimization method, so that the technical problem of lower optimization efficiency of image quality can be solved. Compared with the prior art, the beneficial effects of the computer readable storage medium provided by the embodiment of the present application are the same as those of the image quality optimization method provided by the first embodiment, and are not described in detail herein.
Example six
The embodiments of the present application also provide a computer program product comprising a computer program which, when executed by a processor, implements the steps of the image quality optimization method as described above.
The computer program product provided by the application can solve the technical problem of low optimization efficiency of image quality. Compared with the prior art, the beneficial effects of the computer program product provided by the embodiment of the present application are the same as those of the image quality optimization method provided by the first embodiment, and are not described in detail herein.
The foregoing description is only of the preferred embodiments of the present application, and is not intended to limit the scope of the application, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein, or any application, directly or indirectly, within the scope of the application.
Claims (10)
1. An image quality optimization method, characterized in that the image quality optimization method comprises:
Obtaining a standard image corresponding to an image to be evaluated, and determining a target image quality evaluation model corresponding to the standard image;
Inputting the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, wherein the evaluation result comprises one of qualification and optimization;
When the evaluation result is to be optimized, generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated, and optimizing the image to be evaluated based on the image optimization strategy.
2. The image quality optimization method according to claim 1, wherein the image optimization strategy is an image processing strategy, and the step of generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated includes:
inquiring an image processing strategy corresponding to the defect type in a preset image processing strategy library;
And adjusting the image parameters of the image to be evaluated based on the image processing strategy to generate a target optimized image.
3. The image quality optimization method of claim 1, wherein prior to the step of inputting the image to be evaluated into the target image quality assessment model, the method further comprises:
Acquiring evaluation labels corresponding to a plurality of training images respectively, and setting each evaluation label and the corresponding training image as an evaluation subtask, wherein the evaluation labels comprise subjective evaluation results corresponding to each training image by a plurality of evaluation staff respectively;
And carrying out iterative training on a preset initial image quality assessment model based on each assessment subtask to obtain a target image quality assessment model.
4. The image quality optimization method according to claim 3, wherein the step of iteratively training a preset initial image quality evaluation model based on each of the evaluation subtasks to obtain a target image quality evaluation model comprises:
acquiring gradients of all the evaluation subtasks;
Screening a plurality of evaluation tasks based on the gradient of each evaluation subtask to obtain a training set, wherein the evaluation tasks comprise a plurality of evaluation subtasks corresponding to each evaluation person;
sequentially selecting an evaluation task from the training set according to a preset sequence to perform iterative training on the initial image quality evaluation model, and acquiring a loss function value of the initial image quality evaluation model in the iterative training process;
and stopping training when the loss function value is converged, and setting the current initial image quality assessment model as a target image quality assessment model.
5. The method of optimizing image quality of claim 4, wherein the step of screening a plurality of evaluation tasks based on a gradient of each of the evaluation subtasks to obtain a training set comprises:
According to gradients of a plurality of evaluation subtasks corresponding to each evaluation personnel, a gradient matrix of the evaluation tasks corresponding to each evaluation personnel is obtained;
determining a similarity value between the gradient matrix of each evaluator and the gradient matrix of other evaluators in the evaluator set;
And screening a preset number of evaluation tasks with front similarity values from the evaluation tasks to form a training set.
6. The image quality optimization method according to any one of claims 1 to 5, wherein the evaluation result further includes non-optimizations, and after the step of obtaining the evaluation result and/or defect type corresponding to the image to be evaluated, the method further includes:
If the evaluation result of the image to be evaluated is that optimization is impossible, inquiring a shooting parameter adjustment strategy corresponding to the defect type in a preset shooting parameter adjustment strategy library based on the defect type corresponding to the image to be evaluated;
And carrying out parameter optimization on initial shooting parameters of shooting equipment corresponding to the image to be evaluated based on the shooting parameter adjustment strategy.
7. An image quality optimizing apparatus, characterized in that the image quality optimizing apparatus comprises:
The model determining module is used for acquiring a standard image corresponding to the image to be evaluated and determining a target image quality evaluation model corresponding to the standard image;
The image evaluation module is used for inputting the image to be evaluated into the target image quality evaluation model to obtain an evaluation result and/or a defect type corresponding to the image to be evaluated, wherein the evaluation result comprises one of qualification and optimization;
And the image optimization module is used for generating a corresponding image optimization strategy according to the defect type corresponding to the image to be evaluated when the evaluation result is to be optimized, and optimizing the image to be evaluated based on the image optimization strategy.
8. An electronic device, the electronic device comprising:
At least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the image quality optimization method of any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a program for realizing the image quality optimization method, the program for realizing the image quality optimization method being executed by a processor to realize the steps of the image quality optimization method according to any one of claims 1 to 6.
10. A computer program product, characterized in that the computer program, when executed by a processor, implements the steps of the image quality optimization method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410448087.2A CN118037602B (en) | 2024-04-15 | 2024-04-15 | Image quality optimization method, device, electronic equipment, medium and program product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410448087.2A CN118037602B (en) | 2024-04-15 | 2024-04-15 | Image quality optimization method, device, electronic equipment, medium and program product |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118037602A true CN118037602A (en) | 2024-05-14 |
| CN118037602B CN118037602B (en) | 2024-07-23 |
Family
ID=90991828
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410448087.2A Active CN118037602B (en) | 2024-04-15 | 2024-04-15 | Image quality optimization method, device, electronic equipment, medium and program product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118037602B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118657702A (en) * | 2024-08-19 | 2024-09-17 | 亚创光电(深圳)有限公司 | Image display processing method, display device, medium and program product |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113870210A (en) * | 2021-09-18 | 2021-12-31 | 广东电网有限责任公司 | An image quality assessment method, device, equipment and storage medium |
| CN114972096A (en) * | 2022-05-31 | 2022-08-30 | Oppo广东移动通信有限公司 | Face image optimization method and device, storage medium and electronic equipment |
| CN115205188A (en) * | 2021-04-13 | 2022-10-18 | 腾讯科技（深圳）有限公司 | Method and related device for evaluating image video quality based on approximation value |
| CN115760822A (en) * | 2022-11-28 | 2023-03-07 | 深圳市捷易科技有限公司 | Image quality detection model establishing method and system |
| CN116091900A (en) * | 2021-10-29 | 2023-05-09 | 华为技术有限公司 | Image processing method, device and storage medium |
| CN117456572A (en) * | 2022-07-18 | 2024-01-26 | 北京字跳网络技术有限公司 | Image quality analysis method, image quality model training method and device |
-
2024
- 2024-04-15 CN CN202410448087.2A patent/CN118037602B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115205188A (en) * | 2021-04-13 | 2022-10-18 | 腾讯科技（深圳）有限公司 | Method and related device for evaluating image video quality based on approximation value |
| CN113870210A (en) * | 2021-09-18 | 2021-12-31 | 广东电网有限责任公司 | An image quality assessment method, device, equipment and storage medium |
| CN116091900A (en) * | 2021-10-29 | 2023-05-09 | 华为技术有限公司 | Image processing method, device and storage medium |
| CN114972096A (en) * | 2022-05-31 | 2022-08-30 | Oppo广东移动通信有限公司 | Face image optimization method and device, storage medium and electronic equipment |
| CN117456572A (en) * | 2022-07-18 | 2024-01-26 | 北京字跳网络技术有限公司 | Image quality analysis method, image quality model training method and device |
| CN115760822A (en) * | 2022-11-28 | 2023-03-07 | 深圳市捷易科技有限公司 | Image quality detection model establishing method and system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118657702A (en) * | 2024-08-19 | 2024-09-17 | 亚创光电(深圳)有限公司 | Image display processing method, display device, medium and program product |
| CN118657702B (en) * | 2024-08-19 | 2025-01-14 | 亚创光电(深圳)有限公司 | Image display processing method, display device, medium and program product |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118037602B (en) | 2024-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10984293B2 (en) | Image processing method and apparatus | |
| CN110288682B (en) | Method and apparatus for controlling changes in a three-dimensional virtual portrait mouth shape | |
| US11663806B2 (en) | Saliency of an object for image processing operations | |
| US20210306559A1 (en) | Photographing methods and devices | |
| CN111833459B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN110062157B (en) | Method and device for rendering image, electronic equipment and computer readable storage medium | |
| CN118037602B (en) | Image quality optimization method, device, electronic equipment, medium and program product | |
| CN110349107B (en) | Image enhancement method, device, electronic equipment and storage medium | |
| CN111199540A (en) | Image quality evaluation method, device, electronic device and storage medium | |
| CN110634163A (en) | Method, device and equipment for calibrating parameters of terminal camera and storage medium | |
| CN114650361B (en) | Shooting mode determining method, shooting mode determining device, electronic equipment and storage medium | |
| CN112036517B (en) | Image defect classification method and device and electronic equipment | |
| CN113938596B (en) | Method and device for capturing images | |
| CN118942092B (en) | Test paper image acquisition method and device, camera, storage medium and product | |
| CN114419473B (en) | Deep learning real-time target detection method based on embedded equipment | |
| CN111314620A (en) | Photographing method and apparatus | |
| CN113238652B (en) | Line of sight estimation method, device, equipment and storage medium | |
| US9451155B2 (en) | Depth-segmenting peak tracking autofocus | |
| CN115460353A (en) | Teaching tracking camera equipment configuration method and device, electronic equipment and medium | |
| CN111385460A (en) | Image processing method and device | |
| CN116248855A (en) | Camera parameter testing method, device, computer equipment and storage medium | |
| CN112435231B (en) | Image quality scale generation method, image quality evaluation method and device | |
| CN115310582A (en) | Method and apparatus for training neural network models | |
| CN111124862A (en) | Intelligent equipment performance testing method and device and intelligent equipment | |
| CN119227770B (en) | Multi-mode large language model quantification method, device, equipment, storage medium and product |
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 |