CN114219808B - Image processing method, apparatus, device, storage medium, and computer program product - Google Patents
Image processing method, apparatus, device, storage medium, and computer program product Download PDFInfo
- Publication number
- CN114219808B CN114219808B CN202111443171.8A CN202111443171A CN114219808B CN 114219808 B CN114219808 B CN 114219808B CN 202111443171 A CN202111443171 A CN 202111443171A CN 114219808 B CN114219808 B CN 114219808B
- Authority
- CN
- China
- Prior art keywords
- image
- resolution
- processing
- initial
- scoring
- 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
- 238000004590 computer program Methods 0.000 title claims abstract description 28
- 238000003672 processing method Methods 0.000 title claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 143
- 238000000034 method Methods 0.000 claims abstract description 24
- 230000000903 blocking effect Effects 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims description 83
- 238000000638 solvent extraction Methods 0.000 claims description 6
- 230000003044 adaptive effect Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 abstract description 7
- 230000000875 corresponding effect Effects 0.000 description 117
- 238000013473 artificial intelligence Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The present application relates to an image processing method, apparatus, device, storage medium and computer program product. The method comprises the following steps: determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image; performing blocking processing on the initial image to obtain a plurality of initial image blocks; inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively; and processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image matched with the target screen resolution. By adopting the method, the image display effect can be improved.
Description
Technical Field
The present application relates to the field of artificial intelligence image recognition technology, and in particular, to an image processing method, apparatus, device, storage medium, and computer program product.
Background
With the development of computer devices, in order to facilitate users to understand images, it is often necessary to display images on a screen, for example, to display business data on the screen in a manner of text description and diagrams, so as to facilitate users to understand business related conditions.
In the conventional technology, the resolution of different screens is different, so that when an image is displayed, the content to be displayed can be reduced or enlarged according to the resolution of the screen. For example, the resolution of the entire image is multiplied by a preset ratio to obtain a scaled resolution, and the image is adjusted to the scaled resolution and then displayed on the screen. However, the scaling method in the conventional technology has the following technical problems: this way of scaling is less flexible and some of the content in the image does not match the screen after scaling, resulting in poor image results being presented.
Disclosure of Invention
In view of the foregoing, it is desirable to provide an image processing method, apparatus, device, computer-readable storage medium, and computer program product that address the above-described problems.
In a first aspect, the present application provides an image processing method. The method comprises the following steps: determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image; performing blocking processing on the initial image to obtain a plurality of initial image blocks; inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively; and processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image matched with the target screen resolution.
In one embodiment, the processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as the image adapted to the target screen resolution includes: obtaining a plurality of candidate image processing schemes based on the resolution range corresponding to the initial image block and the target screen resolution; the candidate image processing scheme comprises adjustment resolutions corresponding to the initial image blocks respectively; processing the initial image based on the candidate image processing scheme to obtain a candidate processing image corresponding to the candidate processing scheme; inputting the candidate processed images into a trained image scoring model for scoring to obtain image scores corresponding to the candidate processed images; and selecting candidate processing images meeting the scoring condition based on the image scoring as target images.
In one embodiment, the training step of the image scoring model includes: acquiring a real image and generating an image; inputting the real image into an image scoring model to be trained for scoring to obtain a first image score corresponding to the real image; inputting the generated image into an image scoring model to be trained for scoring to obtain a second image score corresponding to the generated image; obtaining a scoring model loss value based on the first image score and the second image score, wherein the scoring model loss value and the first image score form a negative correlation, and the scoring model loss value and the second image score form a positive correlation; and adjusting model parameters of the image scoring model to be trained based on the scoring model loss value to obtain a trained image scoring model.
In one embodiment, the scoring condition includes at least one of the following conditions: the ranking of the image scores is prior to the ranking threshold or the image scores are greater than the scoring threshold.
In one embodiment, the training step of the resolution identification model comprises: acquiring a training image and a label image corresponding to the training image; partitioning the tag image based on the resolution of the tag image to obtain a plurality of tag image blocks; acquiring a training image block corresponding to the label image block in the training image; inputting the training image block into a resolution recognition model to be trained for processing to obtain a prediction resolution range corresponding to the training image block; determining a resolution model loss value based on a prediction resolution range corresponding to the training image block and a resolution range corresponding to the label image block; and carrying out parameter adjustment on the resolution recognition model to be trained based on the resolution model loss value to obtain a trained resolution recognition model.
In one embodiment, the performing the blocking processing on the initial image to obtain a plurality of initial image blocks includes: performing object recognition on the initial image to obtain a display object contained in the initial image; and carrying out blocking processing on the initial image based on the display object to obtain a plurality of initial image blocks.
In a second aspect, the present application also provides an image processing apparatus. The device comprises: the resolution and image acquisition module is used for determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image; the block processing module is used for carrying out block processing on the initial image to obtain a plurality of initial image blocks; the resolution range obtaining module is used for inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively; and the target image obtaining module is used for processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image adaptive to the target screen resolution.
In one embodiment, the target image obtaining module is configured to: obtaining a plurality of candidate image processing schemes based on the resolution range of the initial image block pair and the target screen resolution; the candidate image processing scheme comprises adjustment resolutions corresponding to the initial image blocks respectively; processing the initial image based on the candidate image processing scheme to obtain a candidate processing image corresponding to the candidate processing scheme; inputting the candidate processed images into a trained image scoring model for scoring to obtain image scores corresponding to the candidate processed images; and selecting candidate processing images meeting the scoring condition based on the image scoring as target images.
In one embodiment, the training step module of the image scoring model is configured to: acquiring a real image and generating an image; inputting the real image into an image scoring model to be trained for scoring to obtain a first image score corresponding to the real image; inputting the generated image into an image scoring model to be trained for scoring to obtain a second image score corresponding to the generated image; obtaining a scoring model loss value based on the first image score and the second image score, wherein the scoring model loss value and the first image score form a negative correlation, and the scoring model loss value and the second image score form a positive correlation; and adjusting model parameters of the image scoring model to be trained based on the scoring model loss value to obtain a trained image scoring model.
In one embodiment, the scoring condition includes at least one of the following conditions: the ranking of the image scores is prior to the ranking threshold or the image scores are greater than the scoring threshold.
In one embodiment, the training module of the resolution identification model is configured to: acquiring a training image and a label image corresponding to the training image; partitioning the tag image based on the resolution of the tag image to obtain a plurality of tag image blocks; acquiring a training image block corresponding to the label image block in the training image; inputting the training image block into a resolution recognition model to be trained for processing to obtain a prediction resolution range corresponding to the training image block; determining a resolution model loss value based on a prediction resolution range corresponding to the training image block and a resolution range corresponding to the label image block; and carrying out parameter adjustment on the resolution recognition model to be trained based on the resolution model loss value to obtain a trained resolution recognition model.
In one embodiment, the block processing module is configured to: performing object recognition on the initial image to obtain a display object contained in the initial image; and carrying out blocking processing on the initial image based on the display object to obtain a plurality of initial image blocks.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor which when executing the computer program performs the steps of: determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image; performing blocking processing on the initial image to obtain a plurality of initial image blocks; inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively; and processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image matched with the target screen resolution.
In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of: determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image; performing blocking processing on the initial image to obtain a plurality of initial image blocks; inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively; and processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image matched with the target screen resolution.
In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the steps of: determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image; performing blocking processing on the initial image to obtain a plurality of initial image blocks; inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively; and processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image matched with the target screen resolution.
The image processing method, the image processing device, the computer equipment, the storage medium and the computer program product have the following technical effects: the method comprises the steps of dividing an initial image into a plurality of image blocks, determining the resolution range corresponding to each initial image block based on a model, and processing different image blocks of the initial image according to the resolution ranges in a self-adaptive mode, so that the processed target image is matched with the resolution of a screen, and the display effect can be improved.
Drawings
FIG. 1 is a diagram of an application environment for an image processing method in one embodiment;
FIG. 2 is a flow chart of an image processing method in one embodiment;
FIG. 3 is an exemplary diagram of an initial image divided into image blocks in one embodiment;
FIG. 4 is a schematic flow chart of processing an initial image based on a resolution range corresponding to an initial image block and a target screen resolution, and using the processed target image as an image adapted to the target screen resolution in one embodiment;
FIG. 5 is a flow chart of a training step of an image scoring model in one embodiment;
FIG. 6 is a block diagram showing the structure of an image processing apparatus in one embodiment;
Fig. 7 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The image processing method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104 or may be located on a cloud or other network server. The terminal 102 may send an initial image to be displayed to the server 104, and the server 104 may obtain the screen resolutions of the display devices according to which display devices the initial image is to be displayed, as target screen resolutions, execute the method according to the embodiment of the present application, obtain target images adapted to each target screen resolution, and send the target images to the corresponding display devices for display. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, and portable wearable devices, where the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.
In one embodiment, a plurality of different target screen resolutions may be stored in the server 104 in advance, that is, the server 104 may generate target images corresponding to the different target screen resolutions in advance, and store the correspondence between the images and the target screen resolutions. In this way, when an image is to be displayed on a certain display device, the server 104 can acquire the screen resolution corresponding to the display device, and select a target image adapted to the screen resolution of the display device based on the correspondence stored in advance.
In one embodiment, the image processing method provided by the embodiment of the application can be executed by a display device.
In one embodiment, as shown in fig. 2, an image processing method is provided, and the method is applied to the server in fig. 1 for illustration, and includes the following steps:
Step S202, determining an initial image to be processed and a target screen resolution corresponding to the initial image.
The initial image may be an image for displaying service data, and the initial image may include different types of content such as charts, characters, and characters. The target screen resolution is a resolution of a screen displaying the initial image, and may be preset or may be a screen resolution of a device to be displayed. The screen resolution refers to the number of pixels displayed on the screen, for example, the screen resolution is 800×1600, which represents 800 pixels in the horizontal direction and 1600 pixels in the vertical direction. There may be a plurality of target screen resolutions, and the content provided by the embodiment of the present application may be executed for each target screen resolution.
In the embodiment of the application, the server can respond to the image processing instruction, acquire the initial image to be processed indicated by the image processing instruction, and acquire the preset target screen resolution.
Step S204, the initial image is subjected to blocking processing, and a plurality of initial image blocks are obtained.
Blocking is a process of dividing an image. The division may be based on a blank area of the image, may be based on a type of content displayed in the image, or may be based on inputting the initial image into an image division model. For example, an image of business data is displayed through different charts, and a blank area exists between the charts, a boundary line can be selected on the blank area to divide the image into a plurality of image blocks, wherein 'a plurality' means at least two.
In the embodiment of the application, the server acquires the initial image, analyzes the initial image, determines the dividing mode of the initial image, and divides the image based on the dividing mode to obtain a plurality of initial image blocks. For example, as shown in fig. 3, one image may be divided into 4 image blocks.
Step S206, inputting the initial image blocks into the trained resolution recognition model for processing, and obtaining resolution ranges corresponding to the initial image blocks respectively.
Wherein the greater the resolution, the more important the content in the image block is represented. The resolution identification model is used for identifying the resolution range of an image block in the image. There are multiple resolutions within a resolution range that can be selected. The resolution identification model determines a resolution range corresponding to each of the initial image blocks.
In the embodiment of the application, the server can input the initial image blocks obtained by dividing one initial image into the resolution identification model together to obtain the resolution range corresponding to each initial image block. The resolution identification model may be a model corresponding to the resolution of the target screen.
In one embodiment, the resolution recognition model is an artificial intelligence model obtained through pre-training, and when the model is trained, a training image and a corresponding image label can be obtained, wherein the image label comprises a resolution corresponding to each training image block in the training image, the resolution corresponding to the training image block is determined based on a value corresponding to the training image block, and the greater the value is, the greater the resolution is. The resolution to which the training image block corresponds is determined, for example, manually based on the value of the training image block in the training image. For another example, an image pair may be obtained, where the image pair includes a training image and a label image corresponding to the training image, and the label image may be an image with a good effect displayed on a display device corresponding to the target screen resolution after the training image is manually adjusted. The resolution of the label image is less than or equal to the target screen resolution. The server may obtain the resolution of the tag image on each image block as the resolution corresponding to the training image block in the image tag. The training image can be input into a resolution recognition model to be trained for respective resolution recognition, a resolution range to which a predicted training image block belongs is obtained, a resolution range corresponding to the training image block in an image tag is determined, and the model loss value is smaller as the difference between the predicted resolution range and the resolution range actually corresponding to the image tag is smaller. For example, a correspondence relationship between the difference and the loss value may be preset, and if the difference differs by 1 range, the loss value is a-b, and if the difference differs by two ranges, the loss value is a. a and b are positive numbers. The server can adjust model parameters of the resolution identification model to be trained towards the direction of decreasing the model loss value, and the trained resolution identification model is obtained.
Step S208, processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image adaptive to the target screen resolution.
Where the resolution of an image refers to the number of pixels in an image block. The resolution of the target image is smaller than or equal to the resolution of the target screen, namely, the number of pixels of the target image in the horizontal direction and the vertical direction is smaller than or equal to the number of pixels corresponding to the resolution of the target screen.
In the embodiment of the application, the server can select one of the resolutions from the resolution range corresponding to the initial image block, and adjust the resolution of the initial image block to the resolution matched with the selected resolution. These adjusted image blocks are then combined according to the position in the initial image to obtain the target image.
In one embodiment, the resolution range output by the model may include a pixel number range in a horizontal direction and a pixel number range in a vertical direction, when the resolution is selected from the range, the resolution range may be selected according to a ratio of the pixel number in the horizontal direction to the pixel number in the vertical direction in the initial image block, so that the ratio of the pixel number in the horizontal direction to the pixel number in the vertical direction is close to the difference between the ratio of the pixel number in the original image block and the ratio of the pixel number in the original image block, and the condition of the approach may be less than a threshold value. For example, assuming that there are 400 pixels in the horizontal direction and 800 pixels in the vertical direction on the initial image block, the ratio of the number of pixels in the horizontal direction to the number of pixels in the vertical direction is 1:2, and when the resolution is selected from the resolution range output by the model, the ratio of the number of pixels selected in the horizontal direction to the number of pixels selected in the vertical direction is better than 1:2, for example, the difference from 1:2 may be smaller than 0.01.
In the image processing method, the initial image can be segmented to obtain a plurality of image blocks, and the resolution range corresponding to each initial image block is determined based on the model, so that different image blocks of the initial image can be processed in a self-adaptive mode according to the resolution ranges, the processed target image is adaptive to the resolution of the screen, and the display effect can be improved.
In one embodiment, as shown in fig. 4, processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and using the processed target image as the image adapted to the target screen resolution includes:
Step S402, obtaining a plurality of candidate image processing schemes based on the resolution range corresponding to the initial image block and the target screen resolution; the candidate image processing scheme comprises adjustment resolutions respectively corresponding to the initial image blocks.
One candidate processing scheme includes an adjusted resolution, referred to as an adjusted resolution, of each initial image block in the initial image. The initial image is processed based on the candidate processing scheme, and the resolution of the resulting target image is less than or equal to the screen resolution. I.e. the adjusted resolution of each initial image block in a candidate image processing scheme is added to obtain a sum of resolutions smaller than the screen resolution. It can be understood that the number of pixel points corresponding to different directions is smaller than the number of pixel points corresponding to the screen resolution when the resolutions are added. For example, assuming that the screen resolution is 800×1600, the number of pixels representing the screen in the horizontal direction is 800, and the number of pixels in the vertical direction is 1600, assuming that the initial image is divided into 4 image blocks in average as shown in fig. 3: A. b, C, and D, the number of pixels in the horizontal direction of a and B, or the number of pixels in the horizontal direction of C and D, is less than 800 in each candidate processing scheme. The number of pixels in the vertical direction of A and C added, or the number of pixels in the vertical direction of B and D added, is less than 1600.
In the embodiment of the application, when a scheme is formed, the server can select one resolution from the resolution range corresponding to each initial image block. The server may acquire a plurality of schemes, for example, 100 schemes, and then discard schemes having a resolution greater than the resolution of the screen among the schemes, with the remaining schemes as candidate processing schemes.
In one embodiment, the resolution range of the model output may include a horizontal pixel count range and a vertical pixel count range, when the candidate processing scheme is selected from the range, the range may be selected according to a ratio of the horizontal pixel count to the vertical pixel count in the initial image block, so that the ratio of the horizontal pixel count to the vertical pixel count in the initial image block is close to the difference between the ratio of the original pixel count in the initial image block and the condition of the approach may be less than a threshold value. For example, assuming that there are 400 pixels in the horizontal direction and 800 pixels in the vertical direction on the initial image block, the ratio of the number of pixels in the horizontal direction to the number of pixels in the vertical direction is 1:2, and when the resolution is selected from the resolution range output by the model, the ratio of the number of pixels selected in the horizontal direction to the number of pixels selected in the vertical direction is better than 1:2, for example, the difference from 1:2 may be smaller than 0.01.
Step S404, processing the initial image based on the candidate image processing scheme to obtain a candidate processing image corresponding to the candidate processing scheme.
For example, the initial image is divided into 4 image blocks: A. b, C, and D, wherein one candidate processing scheme is that the resolution corresponding to a is 100×200, the resolution corresponding to b is 300×400, the resolution corresponding to C is 100×300, the resolution corresponding to D is 200×200, if the original resolution of a is greater than 100×200, compressing a so that the resolution is 100×200, and similarly, processing B, C and D, and combining A, B, C and D obtained by processing into a candidate processing image corresponding to the candidate processing scheme.
Step S406, inputting the candidate processing images into an image scoring model for scoring, and obtaining image scores corresponding to the candidate processing images.
The image scoring model may be an artificial intelligence model for evaluating the presentation quality of the image or a strategy for manually determining the decision of the presentation quality. The higher the score, the better the representative quality. For example, the image scoring model is an image artificial intelligence model for evaluating whether the image is real, and the scoring model may be, for example, a discrimination model in generating a countermeasure model. The discrimination model may be used to discriminate whether an image is a naturally captured image or a machine-based algorithmic synthesis. If the strategy for judging the display quality is determined manually, the strategy can be to determine the difference between the proportion of the pixel points of each image block in the candidate processing image in the horizontal direction and the proportion of the pixel points of the image block in the vertical direction and the proportion of the pixel points of the image block in the initial image, and if the difference is smaller, the score is higher.
In the embodiment of the application, candidate processing images are respectively input into an image scoring model for processing, and the image scoring model performs quality scoring on the candidate processing images to obtain image scores corresponding to the candidate processing images.
Step S408, selecting the candidate processed image satisfying the scoring condition as the target image based on the image score.
In the embodiment of the application, the candidate processing image can be selected only based on the scoring condition, and the candidate processing image can also be selected by combining other conditions, such as combining the resolution condition. The server may set, as the target image, the candidate processed image satisfying the flat scoring condition and the resolution condition. The resolution condition may be, for example, that the resolution ordering follows a preset ordering in which the resolutions are ordered from large to small so that the resolution of the resulting target image is as small as possible.
In one embodiment, the scoring condition includes at least one of the following conditions: the ranking of the image scores is prior to the ranking threshold or the image scores are greater than the scoring threshold. For example, among all the candidate processed images, the image having the largest image score is set as the target image. The ranking threshold and scoring threshold may be set as desired. For example, the ranking threshold may be 2 and the scoring threshold may be 80. The image scores are ordered in order of magnitude. I.e., the higher the score, the earlier the ranking.
In one embodiment, each candidate processed image may be transmitted to the terminal, and the terminal receives a selection operation from the user, and uses the candidate processed image selected by the user as a target image adapted to the target screen resolution. I.e. forming a plurality of candidate processing schemes, acquiring processed images corresponding to the candidate processing schemes, and manually selecting the processed images. Since the resolution in the candidate processing scheme is within the resolution range identified by the resolution model, the candidate processing scheme is relatively less and more suitable, so that the efficiency of manual selection can be improved, and a suitable image can be selected.
In the embodiment of the application, the resolution of the image block is selected according to the resolution range to form a candidate processing scheme, and then the content scoring model is favored to score, and the resolution range is obtained by the trained resolution recognition model, so that the accuracy is higher but the selection of the resolution is not completely limited, and the image with the quality meeting the requirement is selected by scoring of the image scoring model, so that the display quality of the image can be improved while the target image is matched with the resolution of the target screen.
The artificial intelligence model may be obtained through supervised training. In one embodiment, as shown in FIG. 5, the training step of the image scoring model includes:
step S502, acquiring a real image and generating an image.
The generated image is an image generated by an artificial intelligence model, for example, an image generated by an image generation model. The real image is not generated by an artificial intelligence model, for example, an image obtained by photographing. When the real image is displayed in the display equipment corresponding to the resolution of the target screen, the quality of the real image is better than that of the generated image. The actual image and the generated image may be preset, for example, may be manually selected.
Step S504, inputting the real image into an image scoring model to be trained for scoring, and obtaining a first image score corresponding to the real image.
For example, the image scoring model is a discrimination model in the generated countermeasure model, and the discrimination value of the image outputted by the discrimination model as the true image is used as the true image corresponding first image score. The higher the discrimination value, the higher the likelihood of representing a true image.
And step S506, inputting the generated image into an image scoring model to be trained for scoring, and obtaining a second image score corresponding to the generated image.
And taking the discrimination value of the image output by the discrimination model as the true image as the second image score corresponding to the generated image. The higher the discrimination value, the higher the likelihood of representing a true image. The lower the discrimination value, the higher the likelihood of representing a true image.
Step S508, obtaining a scoring model loss value based on the first image score and the second image score.
The model loss value and the first image score form a negative correlation relationship, namely, the higher the first image score is, the better the model is capable of identifying the real image as the real image, so the smaller the model loss value is. The model loss value is positively correlated with the second image score. I.e. the higher the second image score, the higher the likelihood that the model will generate an image error as a true image, the worse the recognition capability, so the larger the loss value of the model.
The model loss value may be calculated using a cross entropy loss value calculation. The server may obtain a first loss value based on the first image score, obtain a second loss value based on the second image score, and add the first loss value to the second loss value to obtain a scoring model loss value.
And step S510, adjusting model parameters of the image scoring model to be trained based on the scoring model loss value to obtain a trained image scoring model.
In the embodiment of the application, the model parameters can be adjusted towards the direction of decreasing the model loss value so that the scoring capacity of the model is higher and higher, wherein the model parameters can be adjusted for a plurality of times, for example, when the training times reach a threshold value or the model loss value is smaller than the threshold value, training is stopped again so as to obtain a trained image scoring model with high accuracy.
In the embodiment of the application, the image scoring model is obtained based on the training of the real image and the generated image, so that the trained image scoring model has good recognition capability of recognizing whether one image is the real image or the generated image, and therefore, the candidate processing image with better quality can be selected from the real image or the generated image to serve as the target image, and the display effect of the target image is improved.
In some embodiments, the training step of the resolution identification model comprises: acquiring a training image and a label image corresponding to the training image; partitioning the tag image based on the resolution of the tag image to obtain a plurality of tag image blocks; acquiring a training image block corresponding to a label image block in a training image; inputting the training image block into a resolution recognition model to be trained for processing to obtain a prediction resolution range corresponding to the training image block; determining a resolution model loss value based on a prediction resolution range corresponding to the training image block and a resolution range corresponding to the tag image block; and carrying out parameter adjustment on the resolution recognition model to be trained based on the resolution model loss value to obtain a trained resolution recognition model.
The label image and the training image are images with the same image content and different resolutions. The label image is adapted to the target screen resolution, while the training image is not. The resolutions of different image blocks in the tag image may be different, and the server may partition the tag image based on the resolution of the tag image, so that the resolution between at least two image blocks in the partitioned tag image blocks is different, and it can be understood that the image area corresponding to one image block is continuous.
The training image is divided according to the division mode of the label image blocks, so that the label image blocks and the training image blocks have a one-to-one correspondence, and the positions corresponding to the image blocks with the correspondence and the image content are the same.
The prediction resolution range is obtained by identifying the current model parameters of the resolution identification model. A resolution model loss value may be determined based on a difference between the predicted resolution range and a resolution range corresponding to the tag image block, wherein the larger the difference, the larger the loss value. For example, a loss value corresponding to each difference may be set, for example, a difference of one range is set, the loss value is c, a difference of two ranges is set, and the loss values are c+d, where c and d are positive numbers. The server may perform parameter adjustment on the resolution recognition model in a direction of reducing the loss value, and use the adjusted model as a trained resolution recognition model, and it may be understood that the model may be obtained through multiple training.
In the embodiment of the application, the model loss value is determined based on the label image of the training image during training, so that the smaller the difference between the resolution range output by the model obtained by training and the resolution range of the label image is, the resolution of the label image is matched with the resolution of the target screen, the resolution range obtained by the resolution recognition model is matched with the resolution of the target screen, and the recognition capability of the resolution recognition model is improved.
In one embodiment, performing a blocking process on the initial image to obtain a plurality of initial image blocks includes: performing object recognition on the initial image to obtain a display object contained in the initial image; and performing blocking processing on the initial image based on the display object to obtain a plurality of initial image blocks.
The display object in the initial image is a complete display unit, which may be a table, a person or a trend graph, for example. When the blocking processing is performed, the blocking is performed based on the display objects, so that each initial image block contains a complete object, for example, one initial image block corresponds to one display object. For example, assuming that there is a table and a portrait in an image, the image is divided into two image blocks: image blocks including tables and image blocks including figures. The initial image is partitioned according to the display object, so that the resolution of the image is determined by taking the display object as a dimension, and the value contained in one display object has consistency, thereby improving the display effect.
In one embodiment, the display objects may be divided into display objects corresponding to display purposes and display objects corresponding to non-display purposes according to the display purpose division. For example, if there is only one person in one drawing, and the purpose of the drawing is to show a change in expression, the display object includes two: a face display object and a non-face display object.
Along with the development of technology, the data visualization large screen becomes a requirement of very fire explosion, but the data large screen has an important problem of screen adaptation, and the method provided by the embodiment of the application can adaptively select corresponding images based on the resolution of the large screen for display, for example, the resolution corresponding to each large screen can be obtained in advance, the image to be displayed in the large screen is obtained, and generally, the resolution is higher than that of the screen, so that different resolution ranges can be allocated to different contents in the image, and the finally obtained target image can be adapted to the resolution corresponding to the large screen and is an image with relatively smaller resolution.
It should be understood that, although the steps in the flowcharts related to the above embodiments are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides an image processing device for realizing the above-mentioned image processing method. The implementation of the solution provided by the apparatus is similar to the implementation described in the above method, so the specific limitation of one or more embodiments of the image processing apparatus provided below may refer to the limitation of the image processing method hereinabove, and will not be repeated herein.
In one embodiment, as shown in fig. 6, there is provided an image processing apparatus including:
the resolution and image acquisition module 602 is configured to determine an initial image to be processed and a target screen resolution corresponding to the initial image, where the target screen resolution is a resolution of a screen displaying the initial image;
The block processing module 604 is configured to perform a block processing on the initial image to obtain a plurality of initial image blocks;
The resolution range obtaining module 606 is configured to input the initial image blocks into a trained resolution recognition model for processing, so as to obtain resolution ranges corresponding to the initial image blocks respectively;
The target image obtaining module 608 is configured to process the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and use the processed target image as an image adapted to the target screen resolution.
In one embodiment, the target image obtaining module is configured to: obtaining a plurality of candidate image processing schemes based on the resolution range of the initial image block pair and the target screen resolution; the candidate image processing scheme comprises adjustment resolutions corresponding to the initial image blocks respectively; processing the initial image based on the candidate image processing scheme to obtain a candidate processing image corresponding to the candidate processing scheme; inputting the candidate processing images into a trained image scoring model for scoring to obtain image scores corresponding to the candidate processing images; and selecting candidate processed images meeting the scoring condition based on the image scoring as target images.
In one embodiment, the training step module of the image scoring model is configured to: acquiring a real image and generating an image; inputting the real image into an image scoring model to be trained for scoring, and obtaining a first image score corresponding to the real image; inputting the generated image into an image scoring model to be trained for scoring, and obtaining a second image score corresponding to the generated image; obtaining a scoring model loss value based on the first image score and the second image score, wherein the scoring model loss value and the first image score form a negative correlation, and the scoring model loss value and the second image score form a positive correlation; and adjusting model parameters of the image scoring model to be trained based on the scoring model loss value to obtain a trained image scoring model.
In one embodiment, the scoring condition includes at least one of the following conditions: the ranking of the image scores is prior to the ranking threshold or the image scores are greater than the scoring threshold.
In one embodiment, the training module of the resolution identification model is to: acquiring a training image and a label image corresponding to the training image; partitioning the tag image based on the resolution of the tag image to obtain a plurality of tag image blocks; acquiring a training image block corresponding to a label image block in a training image; inputting the training image block into a resolution recognition model to be trained for processing to obtain a prediction resolution range corresponding to the training image block; determining a resolution model loss value based on a prediction resolution range corresponding to the training image block and a resolution range corresponding to the tag image block; and carrying out parameter adjustment on the resolution recognition model to be trained based on the resolution model loss value to obtain a trained resolution recognition model.
In one embodiment, the block processing module is configured to: performing object recognition on the initial image to obtain a display object contained in the initial image; and performing blocking processing on the initial image based on the display object to obtain a plurality of initial image blocks.
The respective modules in the above-described image processing apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above 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. 7. 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 includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used to store resolution data as well as picture data. 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 an image processing method.
It will be appreciated by those skilled in the art that the structure shown in FIG. 7 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of: determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image; performing blocking processing on the initial image to obtain a plurality of initial image blocks; inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively; and processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image matched with the target screen resolution.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of: determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image; performing blocking processing on the initial image to obtain a plurality of initial image blocks; inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively; and processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image matched with the target screen resolution.
In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of: determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image; performing blocking processing on the initial image to obtain a plurality of initial image blocks; inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively; and processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image matched with the target screen resolution.
The user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or sufficiently authorized by each party.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.
Claims (13)
1. An image processing method, the method comprising:
determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image;
performing blocking processing on the initial image to obtain a plurality of initial image blocks;
Inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively;
Processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image adaptive to the target screen resolution;
the processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image adapted to the target screen resolution includes:
Obtaining a plurality of candidate image processing schemes based on the resolution range corresponding to the initial image block and the target screen resolution; the candidate image processing scheme comprises adjustment resolutions corresponding to the initial image blocks respectively;
processing the initial image based on the candidate image processing scheme to obtain a candidate processing image corresponding to the candidate image processing scheme;
inputting the candidate processed images into a trained image scoring model for scoring to obtain image scores corresponding to the candidate processed images;
And selecting candidate processing images meeting the scoring condition based on the image scoring as target images.
2. The method of claim 1, wherein the training of the image scoring model comprises:
Acquiring a real image and generating an image;
Inputting the real image into an image scoring model to be trained for scoring to obtain a first image score corresponding to the real image;
Inputting the generated image into an image scoring model to be trained for scoring to obtain a second image score corresponding to the generated image;
obtaining a scoring model loss value based on the first image score and the second image score, wherein the scoring model loss value and the first image score form a negative correlation, and the scoring model loss value and the second image score form a positive correlation;
And adjusting model parameters of the image scoring model to be trained based on the scoring model loss value to obtain a trained image scoring model.
3. The method of claim 1, wherein the scoring condition comprises at least one of the following conditions: the ranking of the image scores is prior to the ranking threshold or the image scores are greater than the scoring threshold.
4. The method of claim 1, wherein the training of the resolution identification model comprises:
acquiring a training image and a label image corresponding to the training image;
Partitioning the tag image based on the resolution of the tag image to obtain a plurality of tag image blocks;
Acquiring a training image block corresponding to the label image block in the training image;
Inputting the training image block into a resolution recognition model to be trained for processing to obtain a prediction resolution range corresponding to the training image block;
Determining a resolution model loss value based on a prediction resolution range corresponding to the training image block and a resolution range corresponding to the label image block;
And carrying out parameter adjustment on the resolution recognition model to be trained based on the resolution model loss value to obtain a trained resolution recognition model.
5. The method of claim 1, wherein the performing the blocking process on the initial image to obtain a plurality of initial image blocks comprises:
Performing object recognition on the initial image to obtain a display object contained in the initial image;
and carrying out blocking processing on the initial image based on the display object to obtain a plurality of initial image blocks.
6. An image processing apparatus, characterized in that the apparatus comprises:
The resolution and image acquisition module is used for determining an initial image to be processed and a target screen resolution corresponding to the initial image, wherein the target screen resolution is the resolution of a screen displaying the initial image;
the block processing module is used for carrying out block processing on the initial image to obtain a plurality of initial image blocks;
the resolution range obtaining module is used for inputting the initial image blocks into a trained resolution recognition model for processing to obtain resolution ranges corresponding to the initial image blocks respectively;
The target image obtaining module is used for processing the initial image based on the resolution range corresponding to the initial image block and the target screen resolution, and taking the processed target image as an image adaptive to the target screen resolution;
the target image obtaining module is used for:
Obtaining a plurality of candidate image processing schemes based on the resolution range corresponding to the initial image block and the target screen resolution; the candidate image processing scheme comprises adjustment resolutions corresponding to the initial image blocks respectively;
processing the initial image based on the candidate image processing scheme to obtain a candidate processing image corresponding to the candidate image processing scheme;
inputting the candidate processed images into a trained image scoring model for scoring to obtain image scores corresponding to the candidate processed images;
And selecting candidate processing images meeting the scoring condition based on the image scoring as target images.
7. The apparatus of claim 6, wherein the training step module of the image scoring model is configured to:
Acquiring a real image and generating an image;
Inputting the real image into an image scoring model to be trained for scoring to obtain a first image score corresponding to the real image;
Inputting the generated image into an image scoring model to be trained for scoring to obtain a second image score corresponding to the generated image;
obtaining a scoring model loss value based on the first image score and the second image score, wherein the scoring model loss value and the first image score form a negative correlation, and the scoring model loss value and the second image score form a positive correlation;
And adjusting model parameters of the image scoring model to be trained based on the scoring model loss value to obtain a trained image scoring model.
8. The apparatus of claim 6, wherein the scoring condition comprises at least one of: the ranking of the image scores is prior to the ranking threshold or the image scores are greater than the scoring threshold.
9. The apparatus of claim 6, wherein the training module of the resolution identification model is to:
acquiring a training image and a label image corresponding to the training image;
Partitioning the tag image based on the resolution of the tag image to obtain a plurality of tag image blocks;
Acquiring a training image block corresponding to the label image block in the training image;
Inputting the training image block into a resolution recognition model to be trained for processing to obtain a prediction resolution range corresponding to the training image block;
Determining a resolution model loss value based on a prediction resolution range corresponding to the training image block and a resolution range corresponding to the label image block;
And carrying out parameter adjustment on the resolution recognition model to be trained based on the resolution model loss value to obtain a trained resolution recognition model.
10. The apparatus of claim 6, wherein the block processing module is to:
Performing object recognition on the initial image to obtain a display object contained in the initial image;
and carrying out blocking processing on the initial image based on the display object to obtain a plurality of initial image blocks.
11. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 5 when the computer program is executed.
12. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 5.
13. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111443171.8A CN114219808B (en) | 2021-11-30 | 2021-11-30 | Image processing method, apparatus, device, storage medium, and computer program product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111443171.8A CN114219808B (en) | 2021-11-30 | 2021-11-30 | Image processing method, apparatus, device, storage medium, and computer program product |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114219808A CN114219808A (en) | 2022-03-22 |
| CN114219808B true CN114219808B (en) | 2024-08-23 |
Family
ID=80699111
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111443171.8A Active CN114219808B (en) | 2021-11-30 | 2021-11-30 | Image processing method, apparatus, device, storage medium, and computer program product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114219808B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111598779A (en) * | 2020-05-14 | 2020-08-28 | Oppo广东移动通信有限公司 | Image super-resolution processing method and device, electronic device and storage medium |
| CN112149756A (en) * | 2020-10-14 | 2020-12-29 | 深圳前海微众银行股份有限公司 | Model training method, image recognition method, device, equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20190119550A (en) * | 2019-10-02 | 2019-10-22 | 엘지전자 주식회사 | Method and apparatus for enhancing image resolution |
| CN113674152B (en) * | 2021-08-03 | 2024-09-06 | Oppo广东移动通信有限公司 | Image processing method, apparatus, electronic device, and computer-readable storage medium |
-
2021
- 2021-11-30 CN CN202111443171.8A patent/CN114219808B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111598779A (en) * | 2020-05-14 | 2020-08-28 | Oppo广东移动通信有限公司 | Image super-resolution processing method and device, electronic device and storage medium |
| CN112149756A (en) * | 2020-10-14 | 2020-12-29 | 深圳前海微众银行股份有限公司 | Model training method, image recognition method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114219808A (en) | 2022-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8442327B2 (en) | Application of classifiers to sub-sampled integral images for detecting faces in images | |
| CN114549913B (en) | Semantic segmentation method and device, computer equipment and storage medium | |
| US20240153271A1 (en) | Method and apparatus for selecting cover of video, computer device, and storage medium | |
| KR102464362B1 (en) | Method and apparatus for processing holographic image | |
| CN114827711B (en) | Image information display method and device | |
| CN114219808B (en) | Image processing method, apparatus, device, storage medium, and computer program product | |
| CN114546206A (en) | Special-shaped screen display method and device, computer equipment and storage medium | |
| US11615507B2 (en) | Automatic content-aware collage | |
| JP7385416B2 (en) | Image processing device, image processing system, image processing method, and image processing program | |
| TW202044199A (en) | Image processing apparatus and image processing method thereof | |
| US11881025B1 (en) | Compound images | |
| WO2018123202A1 (en) | Moving-image processing device, display device, moving-image processing method, and control program | |
| KR20160039091A (en) | Method of processing image data and display system for display power reduction | |
| KR20230062462A (en) | Method and apparatus for creating a natural three-dimensional digital twin through verification of a reference image | |
| CN116597293A (en) | Multi-mode scene recognition method, device, computer equipment and storage medium | |
| CN111814818B (en) | Display device and image processing method thereof | |
| US20210065332A1 (en) | Content aware image fitting | |
| CN114185630B (en) | Screen recording method, device, computer equipment and storage medium | |
| CN117197086A (en) | Image detection method, device, computer equipment and storage medium | |
| CN114399497B (en) | Text image quality detection method, device, computer equipment and storage medium | |
| CN116360905B (en) | Tablet personal computer device based on ink screen, control method, system and medium | |
| CN111383165B (en) | Image processing method, system and storage medium | |
| CN118260443A (en) | Text drawing method, text drawing device, computer equipment and storage medium | |
| CN114742704A (en) | Image amplification method and model training method based on image amplification | |
| CN117422671A (en) | Medical image detection method, system, storage medium and electronic equipment |
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 |