CN109003231A - Image enhancement method and device and display equipment - Google Patents
Image enhancement method and device and display equipment Download PDFInfo
- Publication number
- CN109003231A CN109003231A CN201810596553.6A CN201810596553A CN109003231A CN 109003231 A CN109003231 A CN 109003231A CN 201810596553 A CN201810596553 A CN 201810596553A CN 109003231 A CN109003231 A CN 109003231A
- Authority
- CN
- China
- Prior art keywords
- image
- layer
- luminance information
- network model
- illumination
- 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
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000005286 illumination Methods 0.000 claims abstract description 72
- 238000003062 neural network model Methods 0.000 claims abstract description 44
- 238000012549 training Methods 0.000 claims description 38
- 238000013527 convolutional neural network Methods 0.000 claims description 30
- 238000001914 filtration Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 5
- 230000001537 neural effect Effects 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 2
- 210000005036 nerve Anatomy 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 7
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003389 potentiating effect Effects 0.000 description 2
- 230000003014 reinforcing effect Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- 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]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The application provides an image enhancement method and device. The image enhancement method provided by the application comprises the following steps: constructing an image enhancement neural network model, wherein the input of the image enhancement neural network model is an image and an illumination layer corresponding to the image, and the output of the image enhancement neural network model is an enhanced image; acquiring an illumination layer corresponding to an image to be processed; and inputting the image to be processed and the illumination layer corresponding to the image to be processed into the image enhancement neural network model, and outputting the enhanced image. According to the image enhancement method and device, the constructed image enhancement neural network model can remove illumination changes in the image to be processed according to the input illumination image layer, and the image enhancement effect is improved.
Description
Technical field
This application involves image processing techniques more particularly to a kind of image enchancing methods, device and display equipment.
Background technique
Currently, usually blackboard writing on the blackboard camera is shot at image, using the image of shooting as teacher in education sector
The notes given lessons, in this way, not only can be convenient teacher by the image of shooting and consulted but also student memory and understanding can be helped.So
And image is possible to be influenced by factors such as blackboard material, ambient light powers during acquisition, and image is caused to occur
Phenomena such as contrast is lower, image information is unobvious, cross-color or boundary information clarity are inadequate, influences the text in image
It reads.Therefore, it is necessary to carry out enhancing processing to image.
With the development of depth learning technology, convolutional neural networks have been widely used in field of image enhancement.Specifically,
By a network end to end, convolution is carried out using a series of low-quality image of convolution kernels to input, is obtained enhanced
Image.However, convolution kernel can not capture in image when carrying out image enhancement using blackboard image of the convolutional neural networks to shooting
Illumination variation, image enhancement effects are poor.
Summary of the invention
In view of this, the application provides a kind of image enchancing method, device and display equipment, to solve to use existing side
Effect poor problem when method enhances blackboard image.
The application first aspect provides a kind of image enchancing method, comprising:
Image enhancement neural network model is constructed, the input of described image strength neural network model is image and the figure
As corresponding illumination figure layer, export as enhanced image；
Obtain the corresponding illumination figure layer of image to be processed；
The image to be processed and the corresponding illumination figure layer of the image to be processed are input to described image enhancing nerve
In network model, enhanced image is exported.
Further, the building image enhancement neural network model, specifically includes:
Convolutional neural networks model is constructed, the input of the convolutional neural networks model is that image and described image are corresponding
Illumination figure layer, the output of the convolutional neural networks model are enhanced image；
Training set is constructed, the training set includes multiple groups training data, and every group of training data includes original image, original graph
As corresponding illumination figure layer and the corresponding enhanced image of original image；
Using the training set training convolutional neural networks model, described image strength neural network model is obtained.
Further, the corresponding illumination figure layer of image is obtained using following methods:
From the first color space conversion it is the second color space comprising the first luminance information by described image, and extracts institute
State the first luminance information of image；
First luminance information is filtered using filtering algorithm, obtains gray level image；
Calculate the average value of the second luminance information of all pixels of the gray level image；
The second luminance information that each pixel of the gray level image is corrected using the average value, obtains illumination figure layer,
Wherein, the second luminance information that the luminance information after each pixel correction is equal to each pixel subtracts the average value.
Further, the filtering algorithm is median filtering algorithm.
Further, the convolutional neural networks model is full convolutional network model.
The application second aspect provides a kind of image intensifier device, comprising: and building module obtains module and processing module,
Wherein,
The building module, for constructing image enhancement neural network model, described image strength neural network model
Input is image and the corresponding illumination figure layer of described image, is exported as enhanced image；
The acquisition module, for obtaining the corresponding illumination figure layer of image to be processed；
The processing module, for the image to be processed and the corresponding illumination figure layer of the image to be processed to be input to
In described image strength neural network model, enhanced image is exported.
Further, the building module is specifically used for building convolutional neural networks model and training set, and described in utilization
The training set training convolutional neural networks model, obtains described image strength neural network model；Wherein, the convolutional Neural
The input of network model is image and the corresponding illumination figure layer of described image, and the output of the convolutional neural networks model is enhancing
Image afterwards；The training set includes multiple groups training data, and every group of training data includes original image, the corresponding light of original image
According to the corresponding enhanced image of figure layer and original image.
Further, the corresponding illumination figure layer of image is obtained using following methods:
From the first color space conversion it is the second color space comprising the first luminance information by described image, and extracts institute
State the first luminance information of image；
First luminance information is filtered using filtering algorithm, obtains gray level image；
Calculate the average value of the second luminance information of all pixels of the gray level image；
The second luminance information that each pixel of the gray level image is corrected using the average value, obtains illumination figure layer,
Wherein, the second luminance information that the luminance information after each pixel correction is equal to each pixel subtracts the average value.
Further, the filtering algorithm is median filtering algorithm.
Further, the convolutional neural networks model is full convolutional network model.
The application third aspect provides a kind of display equipment, including memory, processor and storage are on a memory and can
The computer program run on a processor, the processor realize times that the application first aspect provides when executing described program
The step of one image enchancing method.
The application fourth aspect provides a kind of computer storage medium, is stored thereon with computer program, described program quilt
Processor realizes the step of any image Enhancement Method that the application first aspect provides when executing.
Image enchancing method, device and display equipment provided by the present application, by constructing image enhancement neural network model,
And the input of the image enhancement neural network model of building is image and the corresponding illumination figure layer of image, is exported as enhanced figure
Picture, in this way, when being enhanced using the model image to be processed, by obtaining the corresponding illumination figure layer of image to be processed, and
Image to be processed and image to be processed are input in the image enhancement neural network model built, in this way, the image increases
Strong neural network model can remove the illumination variation in image to be processed according to the illumination figure layer of input, improve the increasing of image
Potent fruit.
Detailed description of the invention
Fig. 1 is the flow chart of the application image enchancing method embodiment one；
Fig. 2 is the flow chart of the building image enhancement neural network model shown in one exemplary embodiment of the application；
Fig. 3 is the flow chart for obtaining the corresponding illumination figure layer of image shown in one exemplary embodiment of the application；
Fig. 4 is the original image shown in one exemplary embodiment of the application and the corresponding illumination of original image got
The schematic diagram of figure layer；
Fig. 5 be by Fig. 4 original image and the corresponding illumination figure layer of original image be input to image enhancement neural network
After model, the schematic diagram of the enhanced image of output；
Fig. 6 is the hardware structure diagram of the image intensifier device place display equipment shown in one exemplary embodiment of the application；
Fig. 7 is the structural schematic diagram of the application image intensifier device embodiment one.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the application.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects be described in detail in claims, the application.
It is only to be not intended to be limiting the application merely for for the purpose of describing particular embodiments in term used in this application.
It is also intended in the application and the "an" of singular used in the attached claims, " described " and "the" including majority
Form, unless the context clearly indicates other meaning.It is also understood that term "and/or" used herein refers to and wraps
It may be combined containing one or more associated any or all of project listed.
It will be appreciated that though various information, but this may be described using term first, second, third, etc. in the application
A little information should not necessarily be limited by these terms.These terms are only used to for same type of information being distinguished from each other out.For example, not departing from
In the case where the application range, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as
One information.Depending on context, word as used in this " if " can be construed to " ... when " or " when ...
When " or " in response to determination ".
The application provides a kind of image enchancing method, device and display equipment, to solve using existing method to blackboard
Effect poor problem when image is enhanced.
Image enchancing method, device and display equipment provided by the present application, by constructing image enhancement neural network model,
And the input of the image enhancement neural network model of building is image and the corresponding illumination figure layer of image, is exported as enhanced figure
Picture, in this way, when being enhanced using the model image to be processed, by obtaining the corresponding illumination figure layer of image to be processed, and
Image to be processed and image to be processed are input in the image enhancement neural network model built, in this way, the image increases
Strong neural network model can remove the illumination variation in image to be processed according to the illumination figure layer of input, improve the increasing of image
Potent fruit.
It is described in detail below with technical aspect of the specific embodiment to the application.These specific implementations below
Example can be combined with each other, and the same or similar concept or process may be repeated no more in some embodiments.
Fig. 1 is the flow chart of the application image enchancing method embodiment one.The executing subject of the present embodiment can be individually
Image intensifier device, be also possible to be integrated with the display equipment of image intensifier device.It is below to be integrated with figure with executing subject
It is illustrated for the display equipment of image intensifying device.Fig. 1 is please referred to, method provided in this embodiment may include:
S101, building image enhancement neural network model, the input of above-mentioned image enhancement neural network model be image and
The corresponding illumination figure layer of above-mentioned image, exports as enhanced image.
Specifically, Fig. 2 is the flow chart of the building image enhancement network model shown in one exemplary embodiment of the application.Please
Referring to Fig. 2, in this step, image enhancement neural network model can be constructed with the following method, method includes the following steps:
S201, building convolutional neural networks model, the input of above-mentioned convolutional neural networks model are image and above-mentioned image
Corresponding illumination figure layer, the output of above-mentioned convolutional neural networks model are enhanced image.
Specifically, the convolutional neural networks model of building can be full convolutional network model.In addition, about building convolution mind
Specific implementation process and realization principle through network model may refer to description in the prior art, and details are not described herein again.
It should be noted that the application building convolutional neural networks model, input for image and above-mentioned image it is corresponding
Illumination figure layer, the output of above-mentioned convolutional neural networks model are enhanced image.In this way, can make up in existing model, roll up
Product core can not capture the problem of image irradiation variation.
S202, building training set, above-mentioned training set includes multiple groups training data, every group of training data include original image,
The corresponding illumination figure layer of original image and the corresponding enhanced image of original image.
Specifically, 30 images can be chosen, and then obtain this corresponding illumination figure layer of 30 images and enhanced
Image.Specifically, will be in following implementation about the concrete methods of realizing and realization principle for obtaining the corresponding illumination figure layer of image
It is discussed in detail in example, details are not described herein again.In addition, this 30 can be obtained by existing image enchancing method in this step
The corresponding enhanced image of image.It further, can also be by every image, the corresponding illumination figure layer of every image and every
It opens the corresponding enhanced image correspondence of image and is divided into image block, for example, 9 pieces of image blocks are divided into, in this way, can obtain
270 groups of training datas.
S203, above-mentioned convolutional neural networks model is trained using above-mentioned training set, obtains above-mentioned image enhancement neural network
Model.
Specifically, the above-mentioned convolutional neural networks model of above-mentioned training set training can be utilized using existing algorithm, obtain
To above-mentioned image enhancement neural network model.For example, above-mentioned convolutional Neural net can be trained using error backpropagation algorithm
Network model.
S102, the corresponding illumination figure layer of image to be processed is obtained.
Specifically, Fig. 3 is the flow chart for obtaining the corresponding illumination figure layer of image shown in one exemplary embodiment of the application.
Referring to figure 3., the illumination figure layer of image to be processed can be obtained using following methods, and this method may include steps of:
S301, by image to be processed from the first color space conversion be the second color space comprising the first luminance information,
And extract the first luminance information of above-mentioned image to be processed.
Specifically, the first color space is red, green, blue (RGB) color space or other color spaces.Second face
The colour space is hue, saturation, intensity (HIS) color space, brightness, coloration (YUV) or other colors comprising luminance information
Space.It is illustrated so that the first color space is RGB color, the second color space is YUV color space as an example below.This
In step, image to be processed is just converted into the YUV color space comprising the first luminance information from RGB color.Specifically,
Image to be processed can be converted into the YUV color space comprising the first luminance information from RGB color according to following formula:
Y=0.30R+0.59G+0.11B U=0.493 (B-Y) V=0.877 (R-Y)
Further, empty when image to be processed is converted to the YUV color comprising the first luminance information from RGB color
Between after, the first luminance information of image to be processed can be extracted.
S302, above-mentioned first luminance information is filtered using filtering algorithm, obtains gray level image.
Specifically, can be filtered using median filtering algorithm to the first luminance information in this step, obtain ash
Spend image.Concrete principle about median filtering algorithm may refer to description in the prior art, and details are not described herein again.
The average value of second luminance information of all pixels of S303, the above-mentioned gray level image of calculating.
Specifically, in this step, just calculate all pixels of above-mentioned gray level image second is bright after obtaining gray level image
Spend the average value of information.For example, gray level image includes 4 pixels, the second luminance information of each pixel isAt this point,
The average value that the second luminance information of all pixels of gray level image can be calculated is a, wherein a=(A+B+C+D)/4.
S304, corrected using above-mentioned average value above-mentioned gray level image each pixel the second luminance information, obtain illumination
Figure layer, wherein the second luminance information that the luminance information after each pixel correction is equal to each pixel subtracts above-mentioned average value.
Specifically, the second luminance information of pixel each in above-mentioned gray level image can be subtracted above-mentioned flat in this step
Mean value, and then obtain illumination figure layer.In conjunction with above example, after amendment, the brightness of each pixel of obtained illumination figure layer is believed
Breath is
Fig. 4 is the original image shown in one exemplary embodiment of the application and the corresponding illumination of original image got
The schematic diagram of figure layer.Referring to figure 4., the A image in Fig. 4 is original image, when process step S1021 and S1022 are to original graph
After processing, gray level image (the B image in Fig. 4) is obtained, further, after step S1023 and S1024, is somebody's turn to do
The illumination figure layer (the C image in Fig. 4) of original image.
S103, above-mentioned image to be processed and the corresponding illumination figure layer of above-mentioned image to be processed are input to above-mentioned image enhancement
In neural network model, enhanced image is exported.
Specifically, Fig. 5 be by Fig. 4 original image and the corresponding illumination figure layer of original image be input to image enhancement mind
After network model, the schematic diagram of the enhanced image of output.As seen from Figure 5, using method provided by the present application to blackboard
After image carries out enhancing processing, reinforcing effect is preferable.
Image enchancing method and device provided by the present application, by constructing image enhancement neural network model, and construct
The input of image enhancement neural network model is image and the corresponding illumination figure layer of image, is exported as enhanced image, in this way,
When being enhanced using the model image to be processed, by obtaining the corresponding illumination figure layer of image to be processed, and will be to be processed
Image and image to be processed are input in the image enhancement neural network model built, in this way, the image enhancement nerve net
Network model can remove the illumination variation in image to be processed according to the illumination figure layer of input, improve the reinforcing effect of image.
Corresponding with the embodiment of aforementioned image enchancing method, present invention also provides the embodiments of image intensifier device.
The embodiment of the application image intensifier device can be using on the display device.Installation practice can pass through software
It realizes, can also be realized by way of hardware or software and hardware combining.Taking software implementation as an example, as on a logical meaning
Device, be to be read computer program instructions corresponding in nonvolatile memory by the processor of display equipment where it
Into memory, operation is formed.For hardware view, as shown in fig. 6, showing equipment where the application image intensifier device
A kind of hardware structure diagram filled in embodiment other than memory 810 shown in fig. 6, processor 820 and network interface 830
Display equipment where setting can also include other hardware, no longer to this generally according to the actual functional capability of the image intensifier device
It repeats.
Referring to FIG. 7, image intensifier device provided by the present application, comprising: building module 910 obtains module 920 and processing
Module 930, wherein
The building module 910, for constructing image enhancement neural network model, described image strength neural network model
Input be image and the corresponding illumination figure layer of described image, export as enhanced image；
The acquisition module 920, for obtaining the corresponding illumination figure layer of image to be processed；
The processing module 930, for the image to be processed and the corresponding illumination figure layer of the image to be processed is defeated
Enter into described image strength neural network model, exports enhanced image.
Further, the building module 910 is specifically used for building convolutional neural networks model and training set, and utilizes
The training set training convolutional neural networks model, obtains described image strength neural network model；Wherein, the convolution
The input of neural network model is image and the corresponding illumination figure layer of described image, the output of the convolutional neural networks model are
Enhanced image；The training set includes multiple groups training data, and every group of training data includes original image, original image correspondence
Illumination figure layer and the corresponding enhanced image of original image.
Further, the corresponding illumination figure layer of image is obtained using following methods:
From the first color space conversion it is the second color space comprising the first luminance information by described image, and extracts institute
State the first luminance information of image；
First luminance information is filtered using filtering algorithm, obtains gray level image；
Calculate the average value of the second luminance information of all pixels of the gray level image；
The second luminance information that each pixel of the gray level image is corrected using the average value, obtains illumination figure layer,
Wherein, the second luminance information that the luminance information after each pixel correction is equal to each pixel subtracts the average value.
Further, the filtering algorithm is median filtering algorithm.
Further, the convolutional neural networks model is full convolutional network model.
Please continue to refer to Fig. 6, the application third aspect also provides a kind of display equipment, including memory 810, processor
820 and storage on a memory and the computer program that can run on a processor, when the execution of processor 820 described program
The step of realizing any image Enhancement Method provided by the present application.
Specifically, the display equipment is in addition to including memory 810 shown in fig. 6, processor 820 and network interface 830
Except, it can also include other hardware, this is repeated no more.
In addition, the display equipment for being adapted for carrying out computer program includes, such as general and/or special microprocessor, or
The central processing unit of any other type.In general, central processing unit will be from read-only memory and/or random access memory
Receive instruction and data.The basic module of display equipment includes central processing unit for being practiced or carried out instruction and is used for
One or more memory devices of store instruction and data.In general, display equipment will also be including one for storing data
Or multiple mass-memory units, such as disk, magneto-optic disk or CD etc., or display equipment will operationally with this large capacity
The coupling of storage equipment is to receive from it data or have both at the same time to its transmission data or two kinds of situations.However, display equipment is not
It is must have such equipment.In addition, display equipment can be embedded in another equipment, such as mobile phone, individual digital
Assistant (PDA), Mobile audio frequency or video player, game console, global positioning system (GPS) receiver or for example general
The portable memory apparatus of universal serial bus (USB) flash drive, names just a few.
The application fourth aspect also provides a kind of computer storage medium, is stored thereon with computer program, described program
The step of any image Enhancement Method provided by the present application is realized when being executed by processor.
Specifically, being suitable for storing computer program instructions and the computer-readable medium of data including the non-of form of ownership
Volatile memory, medium and memory devices, for example including semiconductor memory devices (such as EPROM, EEPROM and flash memory
Equipment), disk (such as internal hard drive or removable disk), magneto-optic disk and CD ROM and DVD-ROM disk.Processor and memory
It by supplemented or can be incorporated in dedicated logic circuit.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (10)
1. a kind of image enchancing method, which is characterized in that the described method includes:
Image enhancement neural network model is constructed, the input of described image strength neural network model is image and described image pair
The illumination figure layer answered exports as enhanced image；
Obtain the corresponding illumination figure layer of image to be processed；
The image to be processed and the corresponding illumination figure layer of the image to be processed are input to described image strength neural network
In model, enhanced image is exported.
2. the method according to claim 1, wherein the building image enhancement neural network model, specific to wrap
It includes:
Convolutional neural networks model is constructed, the input of the convolutional neural networks model is image and the corresponding illumination of described image
Figure layer, the output of the convolutional neural networks model are enhanced image；
Training set is constructed, the training set includes multiple groups training data, and every group of training data includes original image, original image pair
The corresponding enhanced image of the illumination figure layer and original image answered；
Using the training set training convolutional neural networks model, described image strength neural network model is obtained.
3. method according to claim 1 or 2, which is characterized in that the corresponding illumination figure layer of image is obtained using following methods
It takes:
From the first color space conversion it is the second color space comprising the first luminance information by described image, and extracts the figure
First luminance information of picture；
First luminance information is filtered using filtering algorithm, obtains gray level image；
Calculate the average value of the second luminance information of all pixels of the gray level image；
The second luminance information that each pixel of the gray level image is corrected using the average value, obtains illumination figure layer, wherein
The second luminance information that luminance information after each pixel correction is equal to each pixel subtracts the average value.
4. according to the method described in claim 3, it is characterized in that, the filtering algorithm is median filtering algorithm.
5. according to the method described in claim 2, it is characterized in that, the convolutional neural networks model is full convolutional network mould
Type.
6. a kind of image intensifier device characterized by comprising building module obtains module and processing module, wherein
The building module, for constructing image enhancement neural network model, the input of described image strength neural network model
For image and the corresponding illumination figure layer of described image, export as enhanced image；
The acquisition module, for obtaining the corresponding illumination figure layer of image to be processed；
The processing module, it is described for the image to be processed and the corresponding illumination figure layer of the image to be processed to be input to
In image enhancement neural network model, enhanced image is exported.
7. device according to claim 6, which is characterized in that the building module is specifically used for building convolutional Neural net
Network model and training set, and using the training set training convolutional neural networks model, obtain described image enhancing nerve
Network model；Wherein, the input of the convolutional neural networks model is image and the corresponding illumination figure layer of described image, the volume
The output of product neural network model is enhanced image；The training set includes multiple groups training data, every group of training data packet
Include the corresponding illumination figure layer of original image, original image and the corresponding enhanced image of original image.
8. device according to claim 6, which is characterized in that the corresponding illumination figure layer of image is obtained using following methods:
From the first color space conversion it is the second color space comprising the first luminance information by described image, and extracts the figure
First luminance information of picture；
First luminance information is filtered using filtering algorithm, obtains gray level image；
Calculate the average value of the second luminance information of all pixels of the gray level image；
The second luminance information that each pixel of the gray level image is corrected using the average value, obtains illumination figure layer, wherein
The second luminance information that luminance information after each pixel correction is equal to each pixel subtracts the average value.
9. a kind of computer storage medium, is stored thereon with computer program, which is characterized in that described program is executed by processor
The step of any one of Shi Shixian claim 1-5 the method.
10. a kind of display equipment including memory, processor and stores the calculating that can be run on a memory and on a processor
Machine program, which is characterized in that the processor realizes the step of any one of claim 1-5 the method when executing described program
Suddenly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810596553.6A CN109003231B (en) | 2018-06-11 | 2018-06-11 | Image enhancement method and device and display equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810596553.6A CN109003231B (en) | 2018-06-11 | 2018-06-11 | Image enhancement method and device and display equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109003231A true CN109003231A (en) | 2018-12-14 |
| CN109003231B CN109003231B (en) | 2021-01-29 |
Family
ID=64601211
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810596553.6A Active CN109003231B (en) | 2018-06-11 | 2018-06-11 | Image enhancement method and device and display equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109003231B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109741281A (en) * | 2019-01-04 | 2019-05-10 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and terminal |
| CN110349107A (en) * | 2019-07-10 | 2019-10-18 | 北京字节跳动网络技术有限公司 | Method, apparatus, electronic equipment and the storage medium of image enhancement |
| CN110428375A (en) * | 2019-07-24 | 2019-11-08 | 东软医疗系统股份有限公司 | A kind of processing method and processing device of DR image |
| WO2020173320A1 (en) * | 2019-02-28 | 2020-09-03 | 腾讯科技（深圳）有限公司 | Image enhancement method and apparatus, and storage medium |
| CN111968188A (en) * | 2020-07-08 | 2020-11-20 | 华南理工大学 | Low-illumination image enhancement processing method, system, device and storage medium |
| CN112069977A (en) * | 2020-09-02 | 2020-12-11 | 腾讯科技（深圳）有限公司 | Image processing method, image processing device, electronic equipment and computer readable medium |
| CN113066019A (en) * | 2021-02-27 | 2021-07-02 | 华为技术有限公司 | Image enhancement method and related device |
| CN113139909A (en) * | 2020-01-19 | 2021-07-20 | 杭州喔影网络科技有限公司 | Image enhancement method based on deep learning |
| WO2022110027A1 (en) * | 2020-11-27 | 2022-06-02 | Boe Technology Group Co., Ltd. | Computer-implemented image-processing method, image-enhancing convolutional neural network, and computer product |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040184657A1 (en) * | 2003-03-18 | 2004-09-23 | Chin-Teng Lin | Method for image resolution enhancement |
| CN106846261A (en) * | 2016-12-21 | 2017-06-13 | 大连海事大学 | Underwater picture processing method based on convolutional neural networks |
| CN107578390A (en) * | 2017-09-14 | 2018-01-12 | 长沙全度影像科技有限公司 | A kind of method and device that image white balance correction is carried out using neutral net |
-
2018
- 2018-06-11 CN CN201810596553.6A patent/CN109003231B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040184657A1 (en) * | 2003-03-18 | 2004-09-23 | Chin-Teng Lin | Method for image resolution enhancement |
| CN106846261A (en) * | 2016-12-21 | 2017-06-13 | 大连海事大学 | Underwater picture processing method based on convolutional neural networks |
| CN107578390A (en) * | 2017-09-14 | 2018-01-12 | 长沙全度影像科技有限公司 | A kind of method and device that image white balance correction is carried out using neutral net |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109741281A (en) * | 2019-01-04 | 2019-05-10 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and terminal |
| WO2020173320A1 (en) * | 2019-02-28 | 2020-09-03 | 腾讯科技（深圳）有限公司 | Image enhancement method and apparatus, and storage medium |
| US11790497B2 (en) | 2019-02-28 | 2023-10-17 | Tencent Technology (Shenzhen) Company Limited | Image enhancement method and apparatus, and storage medium |
| CN110349107A (en) * | 2019-07-10 | 2019-10-18 | 北京字节跳动网络技术有限公司 | Method, apparatus, electronic equipment and the storage medium of image enhancement |
| CN110428375A (en) * | 2019-07-24 | 2019-11-08 | 东软医疗系统股份有限公司 | A kind of processing method and processing device of DR image |
| CN110428375B (en) * | 2019-07-24 | 2024-03-01 | 东软医疗系统股份有限公司 | DR image processing method and device |
| CN113139909A (en) * | 2020-01-19 | 2021-07-20 | 杭州喔影网络科技有限公司 | Image enhancement method based on deep learning |
| CN111968188B (en) * | 2020-07-08 | 2023-08-22 | 华南理工大学 | Low-light image enhancement processing method, system, device and storage medium |
| CN111968188A (en) * | 2020-07-08 | 2020-11-20 | 华南理工大学 | Low-illumination image enhancement processing method, system, device and storage medium |
| CN112069977A (en) * | 2020-09-02 | 2020-12-11 | 腾讯科技（深圳）有限公司 | Image processing method, image processing device, electronic equipment and computer readable medium |
| CN112069977B (en) * | 2020-09-02 | 2024-09-06 | 腾讯科技（深圳）有限公司 | Image processing method, device, electronic equipment and computer readable medium |
| WO2022110027A1 (en) * | 2020-11-27 | 2022-06-02 | Boe Technology Group Co., Ltd. | Computer-implemented image-processing method, image-enhancing convolutional neural network, and computer product |
| CN113066019A (en) * | 2021-02-27 | 2021-07-02 | 华为技术有限公司 | Image enhancement method and related device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109003231B (en) | 2021-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109003231A (en) | Image enhancement method and device and display equipment | |
| US8290252B2 (en) | Image-based backgrounds for images | |
| CN103517046B (en) | Method, apparatus and computer program product for capturing video content | |
| CN101742340B (en) | Image optimization editing method and device | |
| JP5847341B2 (en) | Image processing apparatus, image processing method, program, and recording medium | |
| CN101621704A (en) | Color enhancement for graphic images | |
| KR20170115503A (en) | Apparatus and method for reverse-ton mapping images | |
| JP2007513434A5 (en) | ||
| CN104380727A (en) | Image processing device and image processing method | |
| US20170039991A1 (en) | Color compensation circuit, display apparatus, and color compensation method | |
| US20140198987A1 (en) | Structure Descriptors for Image Processing | |
| CN109427041A (en) | A kind of image white balance method and system, storage medium and terminal device | |
| JP2022550008A (en) | Parallel Histogram Computation by Application to Palette Table Derivation | |
| CN109685750A (en) | Image enchancing method and calculating equipment | |
| CN102377911A (en) | Image processing apparatus, image processing method, and camera module | |
| CN113052768A (en) | Method for processing image, terminal and computer readable storage medium | |
| WO2017143812A1 (en) | Method and device for differentiating objects | |
| US8509529B2 (en) | Color-image representative color decision apparatus and method of controlling operation thereof | |
| US8498332B2 (en) | Chroma supression features | |
| CN104243942B (en) | Image processing method and image processing apparatus | |
| CN108235024B (en) | Method and device for compressing image | |
| CN114202486B (en) | Mosaic removal method and system for capsule endoscope image | |
| WO2022052820A1 (en) | Data processing method, system, and apparatus | |
| CN104954767B (en) | A kind of information processing method and electronic equipment | |
| US7305126B2 (en) | Photo dazzle color temperature correction |
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 |