CN106550244A - The picture quality enhancement method and device of video image - Google Patents
The picture quality enhancement method and device of video image Download PDFInfo
- Publication number
- CN106550244A CN106550244A CN201510590614.4A CN201510590614A CN106550244A CN 106550244 A CN106550244 A CN 106550244A CN 201510590614 A CN201510590614 A CN 201510590614A CN 106550244 A CN106550244 A CN 106550244A
- Authority
- CN
- China
- Prior art keywords
- target image
- region
- frequency information
- image
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 81
- 238000001914 filtration Methods 0.000 claims abstract description 66
- 230000002708 enhancing effect Effects 0.000 claims abstract description 31
- 238000012545 processing Methods 0.000 claims abstract description 25
- 230000002146 bilateral effect Effects 0.000 claims description 37
- 238000003708 edge detection Methods 0.000 claims description 7
- 230000002093 peripheral effect Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 5
- 238000000926 separation method Methods 0.000 abstract description 4
- 238000005728 strengthening Methods 0.000 abstract description 3
- 230000006835 compression Effects 0.000 description 3
- 238000007906 compression Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005755 formation reaction Methods 0.000 description 1
- 238000000265 homogenisation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
- H04N21/2343—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving reformatting operations of video signals for distribution or compliance with end-user requests or end-user device requirements
- H04N21/234354—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving reformatting operations of video signals for distribution or compliance with end-user requests or end-user device requirements by altering signal-to-noise ratio parameters, e.g. requantization
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
Abstract
This application discloses a kind of picture quality enhancement method and device of video image, which arranges different enhancing yardsticks for different types of image-region in advance, for each two field picture of target video, the separation of its high-frequency information and low-frequency information is realized by Filtering Processing, and the preset kind region included in recognizing the two field picture, and then the high-frequency information to different preset kind regions carries out enhancement process using the corresponding default yardstick that strengthens, most enhanced high-frequency information is superimposed with corresponding low-frequency information at last, obtains the image after picture quality enhancement.Relative to existing coherence enhancing mode, the embodiment of the present application adopts multi-scale enhancement mode, the different enhancing that zones of different in target image can be met requires, avoid strengthening excessively and strengthen not enough phenomenon, so that the reinforced effects in each region reach most preferably, so as to the overall image quality for lifting video image.
Description
Technical field
The application is related to technical field of image processing, more particularly to a kind of picture quality enhancement method and device of video image.
Background technology
Internet video be in recent years one develop very fast industry, play in the life of people more and more important
Role.In order to the bandwidth with different user is adapted, original video is transcoded into different code checks and is supplied by the Web Video Service chamber of commerce
User selects, not smooth so as to avoid the Internet video of high code check from playing under the network environment of little bandwidth.
Above-mentioned video code conversion process is actually also a kind of video compression, i.e., the net relative to original video, after transcoding
Can all there is different degrees of compression in network video, cause video image sharpness to subtract greatly, image detail disappears in a large number, picture
The phenomenons such as distortion.Therefore, prior art generally carries out picture quality enhancement process to video image while Internet video is played,
With the high-frequency information for recovering loss as far as possible, the broadcasting experience of user is lifted.
The ultimate principle of above-mentioned picture quality enhancement is to be analyzed by each two field picture to video to be played in real time, determines image
In color information, the data such as object edge, by strengthening algorithm above-mentioned data are carried out with the enhancing of some scale, with sharp
Compound body edge sharpening, lifts contrast, saturation etc. so that image quality gets a promotion.But, prior art is general
Lead to identical for the enhancing yardstick of coherence enhancing method, i.e. video image each several part, so easily lead to some parts and increase
It is strong excessively, and then difference increase between consecutive frame, seriality are reduced, cause to play flicker；And for enhancing not
Foot part, then enhanced image quality cannot still meet requirement.
The content of the invention
To overcome problem present in correlation technique, the application to provide a kind of picture quality enhancement method and device of video image.
The application first aspect provides a kind of picture quality enhancement method of video image, including：
Each two field picture of target video is obtained successively as target image, and Filtering Processing is performed to the target image, institute is obtained
State the low-frequency information and high-frequency information of target image；
Recognize the preset kind region in the target image；
According to the corresponding default enhancing yardstick in each preset kind region, the height to respective regions in the target image respectively
Frequency information carries out enhancement process；
High-frequency information after enhancement process is superimposed with corresponding low-frequency information, the target figure after picture quality enhancement is obtained
Picture.
It is default in the target image with reference in a first aspect, in first aspect in the first feasible embodiment, recognizing
Type area, including：
Whether there is at least one in following preset kind region in recognizing the target image：It is Mosaic style region, straight
Line peripheral type region and people's shape of face region.
With reference to the first feasible embodiment of first aspect, in second feasible embodiment of first aspect, identification
With the presence or absence of at least one in following preset kind region in the target image：Mosaic style region, linear edge type
Region and people's shape of face region, including：
Recognized based on edge detection method and whether there is in the target image Mosaic style region；
Recognized based on Hough transform and whether there is in the target image linear edge type region；
Recognized based on Viola-Jones methods and whether there is in the target image people's shape of face region.
With reference in a first aspect, or first aspect the first feasible embodiment, or second of first aspect is feasible
Embodiment, in first aspect in the third feasible embodiment, according to the corresponding default increasing in each preset kind region
Strong yardstick, carries out enhancement process to the high-frequency information of respective regions in the target image respectively, including：
By formulaHigh frequency in the target image is believed
Breath carries out enhancement process；
Wherein, hkP () is the corresponding original high-frequency information of pixel p,For the corresponding enhanced high frequency letter of pixel p
Breath, s is that pixel p region correspondence is default strengthens yardstick.
With reference in a first aspect, or first aspect the first feasible embodiment, or second of first aspect is feasible
Embodiment, in the 4th kind of feasible embodiment of first aspect, performs Filtering Processing to the target image, including：
Filtering Processing is performed to the target image based on bilateral filtering method.
With reference to the 4th kind of feasible embodiment of first aspect, in the 5th kind of feasible embodiment of first aspect, it is based on
Bilateral filtering method performs Filtering Processing to the target image, including：
If the target image is the first two field picture of target video, directly the target image is performed at bilateral filtering
Reason；
If the target image is not the first two field picture of target video, according to the target image and previous frame image it
Between diversity factor to the target image perform bilateral filtering process.
A kind of the application second aspect, there is provided the picture quality enhancement device of video image, including：Filter unit, recognition unit,
Enhancement unit and superpositing unit；
The filter unit is used for, and obtains each two field picture of target video successively as target image, the target image is held
Row Filtering Processing, obtains the low-frequency information and high-frequency information of the target image；
The recognition unit is used for, and recognizes the preset kind region in the target image；
The enhancement unit is used for, according to the corresponding default enhancing yardstick in each preset kind region, respectively to the target
In image, the high-frequency information of respective regions carries out enhancement process；
The superpositing unit is used for, and the high-frequency information after enhancement process is superimposed with corresponding low-frequency information, picture is obtained
The enhanced target image of matter.
With reference to second aspect, in second aspect in the first feasible embodiment, the recognition unit is specifically configured to：
Whether there is at least one in following preset kind region in recognizing the target image：It is Mosaic style region, straight
Line peripheral type region and people's shape of face region.
The first feasible embodiment with reference to second aspect, it is in second feasible embodiment of second aspect, described
Recognition unit includes：
Mosaic identification module, whether there is Mosaic style area for recognizing based on edge detection method in the target image
Domain；
Linear edge identification module, whether there is linear edge for recognizing based on Hough transform in the target image
Type region；
Face recognition module, whether there is people's shape of face area for recognizing based on Viola-Jones methods in the target image
Domain.
With reference to second aspect, or second aspect the first feasible embodiment, or second of second aspect is feasible
Embodiment, in second aspect in the third feasible embodiment, the filter unit includes：
Bilateral filtering unit, for performing Filtering Processing to the target image based on bilateral filtering method.
The third feasible embodiment with reference to second aspect, it is in the 4th kind of feasible embodiment of second aspect, described
Bilateral filtering unit is specifically configured to：
If the target image is the first two field picture of target video, directly the target image is performed at bilateral filtering
Reason；If the target image is not the first two field picture of target video, according to the target image and previous frame image it
Between diversity factor to the target image perform bilateral filtering process.
From above technical scheme, the embodiment of the present application arranges different enhancing chis for different types of image-region in advance
Degree, for each two field picture of target video, realizes the separation of its high-frequency information and low-frequency information by Filtering Processing, and knows
The preset kind region not included in the two field picture, and then the high-frequency information to different preset kind regions adopts corresponding
The default yardstick that strengthens carries out enhancement process, and most enhanced high-frequency information is superimposed with corresponding low-frequency information at last, obtains picture
The enhanced image of matter.Relative to existing coherence enhancing mode, the embodiment of the present application adopts multi-scale enhancement mode,
The different enhancing that zones of different in target image can be met requires, it is to avoid strengthen excessively and strengthen not enough phenomenon, so as to
The reinforced effects in each region are made to reach most preferably, so as to the overall image quality for lifting video image.
It should be appreciated that the general description of the above and detailed description hereinafter are only exemplary and explanatory, can not
Limit the application.
Description of the drawings
During accompanying drawing herein is merged in description and the part of this specification is constituted, show the enforcement for meeting the present invention
Example, and be used for explaining the principle of the present invention together with description.
Fig. 1 is a kind of flow chart of the picture quality enhancement method of video image that the embodiment of the present application is provided.
Fig. 2 is a kind of structured flowchart of the picture quality enhancement device of video image that the embodiment of the present application is provided.
Fig. 3 is the structured flowchart of the picture quality enhancement device of another kind of video image that the embodiment of the present application is provided.
Specific embodiment
Here in detail exemplary embodiment will be illustrated, its example is illustrated in the accompanying drawings.Explained below is related to attached
During figure, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary is implemented
Embodiment described in example does not represent all embodiments consistent with the present invention.Conversely, they be only with such as
The example of consistent apparatus and method in terms of some described in detail in appended claims, the present invention.
A kind of flow chart of the picture quality enhancement method of video image that Fig. 1 is provided for the embodiment of the present application.The method is especially fitted
For the picture quality enhancement to Internet video.As shown in figure 1, the method is comprised the following steps.
Each two field picture of S11, successively acquisition target video performs Filtering Processing to the target image as target image,
Obtain the low-frequency information and high-frequency information of the target image.
Applicant's research finds that the video to Internet video, particularly low bit- rate can be presented relative to original video bright
The trend of degree homogenization, contrast are reduced, edge blurry, and also many details are also all smudgy, and these problems are all
From the high-frequency information part of video image, therefore want to lift image quality, it is necessary to which these high-frequency informations are processed.There is mirror
In this, high-frequency information is extracted by the present embodiment first by Filtering Processing.Filtering Processing is performed to the i-th two field picture,
What is directly obtained is image information fiIn low-frequency information fi B, corresponding high-frequency information hiCan be calculated by equation below
Obtain：hi=fi-fi B。
S12, the preset kind region recognized in the target image.
S13, according to the corresponding default enhancing yardstick in each preset kind region, respectively to respective area in the target image
The high-frequency information in domain carries out enhancement process.
For example, required according to actual picture quality enhancement, the first preset kind region, the second preset kind region and the 3rd preset
The corresponding default enhancing yardstick of type area is respectively m1, m2 and m3；Learnt by the identification of step S12, target figure
As in, region a1 is the first preset kind region, and region a2 is the second preset kind region, region a3 is the 3rd default class
Type region, then, when enhancement process is carried out to high-frequency information, for m1, region a2 is adopted the enhancing yardstick that region a1 is adopted
Enhancing yardstick be m2, the enhancing yardstick adopted by region a3 is for m3.
S14, the high-frequency information after enhancement process is superimposed with corresponding low-frequency information, obtains the mesh after picture quality enhancement
Logo image.
From step S14, enhanced i-th two field picture is represented by：Wherein,To strengthen
The high-frequency information of the i-th two field picture after process.
From above technical scheme, the embodiment of the present application arranges different enhancing chis for different types of image-region in advance
Degree, for each two field picture of target video, realizes the separation of its high-frequency information and low-frequency information by Filtering Processing, and knows
The preset kind region not included in the two field picture, and then the high-frequency information to different preset kind regions adopts corresponding
The default yardstick that strengthens carries out enhancement process, and most enhanced high-frequency information is superimposed with corresponding low-frequency information at last, obtains picture
The enhanced image of matter.Relative to existing coherence enhancing mode, the embodiment of the present application adopts multi-scale enhancement mode,
The different enhancing that zones of different in target image can be met requires, it is to avoid strengthen excessively and strengthen not enough phenomenon, so as to
The reinforced effects in each region are made to reach most preferably, so as to the overall image quality for lifting video image.
In a feasible embodiment, above-mentioned preset kind region can include following at least one：Mosaic style area
Domain, linear edge type region and people's shape of face region；Accordingly, above-mentioned steps S12 are specially：Recognize the target image
In with the presence or absence of at least one in following preset kind region：Mosaic style region, linear edge type region and people's shape of face
Region.
Wherein, mosaic area is generally present in large-area smooth region, mainly because believing to high frequency during video code conversion
The excess compression of breath and produce；If strengthening excessive, in causing image, mosaic effect becomes apparent from.Linear edge type area
Domain mainly for image in sharp keen linear edge, if it is incorrect to strengthen yardstick, image can be caused to lose shape problem.Face
Type region if enhancing yardstick is incorrect, can reduce figure mainly for the face in image generally as the main body of image
The overall visual effect of picture.Therefore, above-mentioned Mosaic style region, linear edge type region and people's shape of face region are to need weight
Point carries out the key area of enhancement process.
In addition, for other non-critical areas in target image, also separately as a kind of type area, using corresponding
The default yardstick that strengthens carries out enhancement process；I.e. through step S12, target image can be divided into into Mosaic style from content
Region, linear edge type region, people's shape of face region and other regions.
In one feasible embodiment of the application, can recognize whether deposit in the target image based on edge detection method
In Mosaic style region.As Mosaic style region is generally the region (such as rectangular area) of regular shape, and easily one
Continuously occur in individual regional area, therefore detection identification can be carried out using simple edge detection method.
In the application in another feasible embodiment, whether can be recognized in the target image based on Hough transform
There is linear edge type region.
In the application in another feasible embodiment, can be recognized in the target image based on Viola-Jones methods
With the presence or absence of people's shape of face region.
Further, since required precision of the embodiment of the present application to region recognition be not high, therefore to improve recognition speed, Ke Yijie
The repeatability between consecutive frame is closed, the identification operation of redundancy is reduced.
In one feasible embodiment of the application, perform in above-mentioned steps S13 and enhancing is performed to the high-frequency information in each region
Process specifically can be according to equation below one：
(formula one)
Wherein, in above-mentioned formula, hiP () is the corresponding original high-frequency information of pixel p,For the corresponding increasing of pixel p
High-frequency information after strong, s is that pixel p region correspondence is default strengthens yardstick.
As a example by still target image described herein above includes Mosaic style, linear edge type, people's shape of face and other regions, phase
Answer, the high-frequency information h of the target imageiConsist ofEnhanced high-frequency information's
Consist ofWherein,
In one feasible embodiment of the application, described in above-mentioned steps S11 to target image perform Filtering Processing,
It is specifically as follows：Filtering Processing is performed to the target image based on bilateral filtering method.
Bilateral filtering is widely used in various graphics processing operations because which can keep marginal information well.This enforcement
Example is specific as follows using the principle that bilateral filtering separates high-frequency information and low-frequency information.
Formula two can be expressed as to the bilateral filtering of pixel x of target image I：
(formula two)
Wherein, NxFor the neighborhood of pixel x,For space weight,For the face of pixel
Colour weight.
In view of two computation complexity of above-mentioned formula is larger, below based on the YUV color spaces that Video coding is commonly used, to public affairs
Formula two is simplified.In YUV color spaces, Y passages represent brightness value, are the main description of image medium-high frequency information
Object, therefore to considering limited brightness value k, i.e. I (x)=k and k ∈ { 0 ..., N-1 } (usual N=255) in filtering；
Then formula two can be converted into equation below three：
(formula three)
Wherein, for each pixel y and each brightness value k, it is defined as follows：And
Jk(y)=Wk(y) * I (y), so as to IBX () can be decomposed into N number of linear filtering response according to N number of value of k, as follows
Formula four：
(formula four)
Above-mentioned linear filtering responseIt is properly termed as image main constituent (the Principle Bilateral Filter of bilateral filtering
Image Component, PBFIC).In practical application, required precision and computation complexity are considered, it is not necessary to
Corresponding PBFIC is calculated to each brightness value k, but all of brightness value [0,255] is divided intoLayer, per layer to take one bright
Angle value L simultaneously asks for its PBFIC, that is, takeCalculate altogetherIndividual PBFIC.Accordingly, target figure
As bilateral filtering value I of pixel x of IBX () can be byWithLinear interpolation is obtained, equation below five：
(formula five)
Wherein,
In fact,The usual very little of value, takeIt is obtained with very high precision.It is above-mentioned double based on what is quantified
Side filtering mode can improve filtering speed to a great extent, and then be conducive to improving the process of whole image enhancing process
Speed, it is ensured that target video smooth playing.
In the application in another feasible embodiment, it is further to improve filtering speed, above-mentioned bilateral filtering method can be with
Further simplified according to the diversity factor between consecutive frame, specifically included：If the target image is target video
First two field picture, then directly perform bilateral filtering to the target image and process, if the target image is not target video
First two field picture, then according to the diversity factor between the target image and previous frame image to the target image perform it is bilateral
Filtering Processing.
Due to there is high similarity between adjacent two field picture, diversity factor very little, therefore, for follow-up two field picture, can
With its intermediate data D corresponding with the similar portion for reusing previous frame image, only to its difference section with previous frame image
Bilateral filtering calculating is carried out, such that it is able to reduce substantial amounts of double counting, filtering speed is improved.Specifically, intermediate data
D can be above-mentioned
For first two field picture (the 0th two field picture), its filter value is calculated according to above-mentioned formula five directly, namely which is low
Frequency information
For i+1 (i ∈ { 0,1,2 ... }) two field picture, its diversity factor with the i-th two field picture is first calculated
To seek exclusive-OR operator), further according to diversity factor Si+1Size determine filter valueAssume that default discrepancy threshold is ε, when
Si+1During ＜ ε, the filter value of i+1 two field picture is to inherit the intermediate data of coming by the i-th two field picture(i.e. i+1 frame figure
As filter value corresponding with the same section of the i-th two field picture), it is not necessary to individually calculate；Work as Si+1During >=ε, i+1 frame figure
The filter value of picture both includedAlso include carrying out bilateral filtering meter to the difference section of i+1 two field picture and the i-th two field picture
The data for obtainingHave：
As seen from the above embodiment, for target video, only bilateral filtering need to be carried out to the full detail of first two field picture and be obtained
Its filter value, i.e. low-frequency information；For follow-up two field picture, with current frame image phase in the filter value of the previous system of battle formations picture of multiplexing
With the corresponding filter value in part, and only the diversity factor at which with previous frame image more than default discrepancy threshold when further calculate
Its filter value corresponding with the difference section of previous frame image.It can be seen that, the present embodiment can reduce substantial amounts of double counting,
Improve filtering speed.In addition, the present embodiment is filtered according to the diversity factor between adjacent two field picture, it can also be ensured that phase
Seriality, temporal consistency between adjacent frame, it is to avoid occur when causing video playback because adjacent two field picture is discontinuous shake,
The problems such as flicker.
A kind of structured flowchart of the picture quality enhancement device of video image that Fig. 2 is provided for the embodiment of the present application.With reference to Fig. 2,
The device includes：Filter unit 100, recognition unit 200, enhancement unit 300 and superpositing unit 400.
The filter unit 100 is configured to, and obtains each two field picture of target video successively as target image, to the target
Image performs Filtering Processing, obtains the low-frequency information and high-frequency information of the target image.
The recognition unit 200 is configured to, and recognizes the preset kind region in the target image.
The enhancement unit 300 is configured to, according to the corresponding default enhancing yardstick in each preset kind region, respectively to institute
The high-frequency information for stating respective regions in target image carries out enhancement process.
The superpositing unit 400 is configured to, and the high-frequency information after enhancement process is superimposed with corresponding low-frequency information,
Obtain the target image after picture quality enhancement.
From above technical scheme, the embodiment of the present application arranges different enhancing chis for different types of image-region in advance
Degree, for each two field picture of target video, realizes the separation of its high-frequency information and low-frequency information by Filtering Processing, and knows
The preset kind region not included in the two field picture, and then the high-frequency information to different preset kind regions adopts corresponding
The default yardstick that strengthens carries out enhancement process, and most enhanced high-frequency information is superimposed with corresponding low-frequency information at last, obtains picture
The enhanced image of matter.Relative to existing coherence enhancing mode, the embodiment of the present application adopts multi-scale enhancement mode,
The different enhancing that zones of different in target image can be met requires, it is to avoid strengthen excessively and strengthen not enough phenomenon, so as to
The reinforced effects in each region are made to reach most preferably, so as to the overall image quality for lifting video image.
In a feasible embodiment, the preset kind region recognized by above-mentioned recognition unit 200 include it is following at least
It is a kind of：Mosaic style region, linear edge type region and people's shape of face region.
The structured flowchart of the picture quality enhancement device of another kind of video image with reference to shown in Fig. 3, above-mentioned recognition unit 200 can
To include：Mosaic identification module 201, linear edge identification module 202 and face recognition module 203.
Wherein, mosaic identification module 201, whether there is in the target image for being recognized based on edge detection method
Mosaic style region；
Linear edge identification module 202, whether there is straight line for recognizing based on Hough transform in the target image
Edge type region；
Face recognition module 203, whether there is face for recognizing based on Viola-Jones methods in the target image
Type region.
In the application in another feasible embodiment, above-mentioned enhancement unit 300 specifically can be by equation below to institute
The high-frequency information stated in target image carries out enhancement process：
Wherein, hkP () is the corresponding original high-frequency information of pixel p,For the corresponding enhanced high frequency letter of pixel p
Breath, s is that pixel p region correspondence is default strengthens yardstick.
Referring now still to Fig. 3, in the application in another feasible embodiment, above-mentioned filter unit can specifically adopt bilateral
Filter unit 101.The bilateral filtering unit is used for, and performs Filtering Processing to the target image based on bilateral filtering method.
More specifically, the bilateral filtering unit is configured to, if the target image is the first two field picture of target video,
Then directly perform bilateral filtering to the target image to process；If the target image is not the first two field picture of target video,
Then bilateral filtering is performed to the target image according to the diversity factor between the target image and previous frame image to process.
With regard to the device in above-described embodiment, wherein modules perform the concrete mode of operation in relevant the method
It has been described in detail in embodiment, explanation will be not set forth in detail herein.
In addition, the embodiment of the present application additionally provides a kind of computer-readable recording medium；Have program stored therein in the storage medium,
When computing device of the program stored in the storage medium by Internet video playback equipment, above-described embodiment can be completed
The part or all of step of described picture quality enhancement method.
Those skilled in the art will readily occur to other of the present invention after considering description and putting into practice invention disclosed herein
Embodiment.The application is intended to any modification, purposes or the adaptations of the present invention, these modifications, purposes
Or adaptations follow the general principle of the present invention and including undocumented in the art known normal of the application
Know or conventional techniques.Description and embodiments are considered only as exemplary, and true scope and spirit of the invention are by under
The claim in face is pointed out.
It should be appreciated that the precision architecture for being described above and being shown in the drawings is the invention is not limited in, and
Various modifications and changes are being carried out without departing from the scope can.The scope of the present invention is limited only by appended claim.
Claims (11)
1. a kind of picture quality enhancement method of video image, it is characterised in that include：
Each two field picture of target video is obtained successively as target image, and Filtering Processing is performed to the target image, institute is obtained
State the low-frequency information and high-frequency information of target image；
Recognize the preset kind region in the target image；
According to the corresponding default enhancing yardstick in each preset kind region, the height to respective regions in the target image respectively
Frequency information carries out enhancement process；
High-frequency information after enhancement process is superimposed with corresponding low-frequency information, the target image after picture quality enhancement is obtained.
2. method according to claim 1, it is characterised in that the preset kind region in the identification target image,
Including：
Whether there is at least one in following preset kind region in recognizing the target image：It is Mosaic style region, straight
Line peripheral type region and people's shape of face region.
3. method according to claim 2, it is characterised in that with the presence or absence of following pre- in the identification target image
If at least one in type area：Mosaic style region, linear edge type region and people's shape of face region, including：
Recognized based on edge detection method and whether there is in the target image Mosaic style region；
Recognized based on Hough transform and whether there is in the target image linear edge type region；
Recognized based on Viola-Jones methods and whether there is in the target image people's shape of face region.
4. the method according to any one of claims 1 to 3, it is characterised in that according to each preset kind region pair
The default enhancing yardstick answered, carries out enhancement process to the high-frequency information of respective regions in the target image respectively, including：
By formula High frequency in the target image is believed
Breath carries out enhancement process；
Wherein, hkP () is the corresponding original high-frequency information of pixel p,For the corresponding enhanced high frequency letter of pixel p
Breath, s is that pixel p region correspondence is default strengthens yardstick.
5. the method according to any one of claims 1 to 3, it is characterised in that filtering is performed to the target image
Process, including：
Filtering Processing is performed to the target image based on bilateral filtering method.
6. method according to claim 5, it is characterised in that the target image is held based on bilateral filtering method
Row Filtering Processing, including：
If the target image is the first two field picture of target video, directly the target image is performed at bilateral filtering
Reason；
If the target image is not the first two field picture of target video, according to the target image and previous frame image it
Between diversity factor to the target image perform bilateral filtering process.
7. the picture quality enhancement device of a kind of video image, it is characterised in that include：Filter unit, recognition unit, enhancing
Unit and superpositing unit；
The filter unit is used for, and obtains each two field picture of target video successively as target image, the target image is held
Row Filtering Processing, obtains the low-frequency information and high-frequency information of the target image；
The recognition unit is used for, and recognizes the preset kind region in the target image；
The enhancement unit is used for, according to the corresponding default enhancing yardstick in each preset kind region, respectively to the target
In image, the high-frequency information of respective regions carries out enhancement process；
The superpositing unit is used for, and the high-frequency information after enhancement process is superimposed with corresponding low-frequency information, picture is obtained
The enhanced target image of matter.
8. device according to claim 7, it is characterised in that the recognition unit is specifically configured to：
Whether there is at least one in following preset kind region in recognizing the target image：It is Mosaic style region, straight
Line peripheral type region and people's shape of face region.
9. device according to claim 8, it is characterised in that the recognition unit includes：
Mosaic identification module, whether there is Mosaic style area for recognizing based on edge detection method in the target image
Domain；
Linear edge identification module, for recognizing the linear edge type that whether there is in the target image based on Hough transform
Region；
Face recognition module, whether there is people's shape of face area for recognizing based on Viola-Jones methods in the target image
Domain.
10. the device according to any one of claim 7 to 9, it is characterised in that the filter unit includes：
Bilateral filtering unit, for performing Filtering Processing to the target image based on bilateral filtering method.
11. devices according to claim 10, it is characterised in that the bilateral filtering unit is specifically configured to：
If the target image is the first two field picture of target video, directly the target image is performed at bilateral filtering
Reason；If the target image is not the first two field picture of target video, according to the target image and previous frame image it
Between diversity factor to the target image perform bilateral filtering process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510590614.4A CN106550244A (en) | 2015-09-16 | 2015-09-16 | The picture quality enhancement method and device of video image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510590614.4A CN106550244A (en) | 2015-09-16 | 2015-09-16 | The picture quality enhancement method and device of video image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106550244A true CN106550244A (en) | 2017-03-29 |
Family
ID=58362770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510590614.4A Pending CN106550244A (en) | 2015-09-16 | 2015-09-16 | The picture quality enhancement method and device of video image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106550244A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107566893A (en) * | 2017-09-19 | 2018-01-09 | 北京奇虎科技有限公司 | Video data handling procedure, device and computing device based on layering |
| CN107633481A (en) * | 2017-09-19 | 2018-01-26 | 北京奇虎科技有限公司 | Image processing method, device, computing device and storage medium based on layering |
| CN108596855A (en) * | 2018-04-28 | 2018-09-28 | 国信优易数据有限公司 | A kind of video image quality Enhancement Method, device and video picture quality enhancement method |
| CN109660821A (en) * | 2018-11-27 | 2019-04-19 | Oppo广东移动通信有限公司 | Method for processing video frequency, device, electronic equipment and storage medium |
| CN110636217A (en) * | 2019-08-29 | 2019-12-31 | 高雪亚 | Automatic field data processing system |
| CN110740253A (en) * | 2019-04-01 | 2020-01-31 | 陈庆梅 | Shooting mode self-adaptive switching method |
| CN111031346A (en) * | 2019-10-28 | 2020-04-17 | 网宿科技股份有限公司 | Method and device for enhancing video image quality |
| WO2021077911A1 (en) * | 2019-10-23 | 2021-04-29 | 华为技术有限公司 | Image flooding processing method and apparatus, and storage medium |
| CN113034412A (en) * | 2021-02-25 | 2021-06-25 | 北京达佳互联信息技术有限公司 | Video processing method and device |
| CN114598902A (en) * | 2022-03-09 | 2022-06-07 | 安徽文香科技有限公司 | Video frame processing method and device and electronic equipment |
| CN115272142A (en) * | 2022-09-30 | 2022-11-01 | 南通市通州区华凯机械有限公司 | Scene image preprocessing method of immersive driving simulator |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030179299A1 (en) * | 2002-03-20 | 2003-09-25 | Hidefumi Okada | Edge emphasizing circuit |
| CN101599171A (en) * | 2008-06-03 | 2009-12-09 | 宝利微电子系统控股公司 | Auto contrast's Enhancement Method and device |
| CN101706954A (en) * | 2009-11-13 | 2010-05-12 | 北京中星微电子有限公司 | Image enhancement method and device thereof as well as image low frequency component computing method and device thereof |
| CN102456223A (en) * | 2010-10-27 | 2012-05-16 | 珠海扬智电子有限公司 | Device and method for carrying out enhancement and scaling on image details |
| CN102915435A (en) * | 2012-10-23 | 2013-02-06 | 哈尔滨工程大学 | Multi-pose face recognition method based on face energy diagram |
| JP2015088918A (en) * | 2013-10-30 | 2015-05-07 | キヤノン株式会社 | Information processing device and information processing method |
| CN104657752A (en) * | 2015-03-17 | 2015-05-27 | 银江股份有限公司 | Deep learning-based safety belt wearing identification method |
-
2015
- 2015-09-16 CN CN201510590614.4A patent/CN106550244A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030179299A1 (en) * | 2002-03-20 | 2003-09-25 | Hidefumi Okada | Edge emphasizing circuit |
| CN101599171A (en) * | 2008-06-03 | 2009-12-09 | 宝利微电子系统控股公司 | Auto contrast's Enhancement Method and device |
| CN101706954A (en) * | 2009-11-13 | 2010-05-12 | 北京中星微电子有限公司 | Image enhancement method and device thereof as well as image low frequency component computing method and device thereof |
| CN102456223A (en) * | 2010-10-27 | 2012-05-16 | 珠海扬智电子有限公司 | Device and method for carrying out enhancement and scaling on image details |
| CN102915435A (en) * | 2012-10-23 | 2013-02-06 | 哈尔滨工程大学 | Multi-pose face recognition method based on face energy diagram |
| JP2015088918A (en) * | 2013-10-30 | 2015-05-07 | キヤノン株式会社 | Information processing device and information processing method |
| CN104657752A (en) * | 2015-03-17 | 2015-05-27 | 银江股份有限公司 | Deep learning-based safety belt wearing identification method |
Non-Patent Citations (3)
| Title |
|---|
| XIAOHUI BIE ET AL: "Enhanced Use of Mattes for Easy Image Composition", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
| 刘丽梅: "HSV空间彩色X线医学图像增强研究", 《云南师范大学学报》 * |
| 朱才高: "红外图像增强算法研究及其DSP实现", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107633481A (en) * | 2017-09-19 | 2018-01-26 | 北京奇虎科技有限公司 | Image processing method, device, computing device and storage medium based on layering |
| CN107566893A (en) * | 2017-09-19 | 2018-01-09 | 北京奇虎科技有限公司 | Video data handling procedure, device and computing device based on layering |
| CN108596855A (en) * | 2018-04-28 | 2018-09-28 | 国信优易数据有限公司 | A kind of video image quality Enhancement Method, device and video picture quality enhancement method |
| CN109660821A (en) * | 2018-11-27 | 2019-04-19 | Oppo广东移动通信有限公司 | Method for processing video frequency, device, electronic equipment and storage medium |
| CN110740253A (en) * | 2019-04-01 | 2020-01-31 | 陈庆梅 | Shooting mode self-adaptive switching method |
| CN110636217A (en) * | 2019-08-29 | 2019-12-31 | 高雪亚 | Automatic field data processing system |
| WO2021077911A1 (en) * | 2019-10-23 | 2021-04-29 | 华为技术有限公司 | Image flooding processing method and apparatus, and storage medium |
| CN111031346A (en) * | 2019-10-28 | 2020-04-17 | 网宿科技股份有限公司 | Method and device for enhancing video image quality |
| CN111031346B (en) * | 2019-10-28 | 2021-11-02 | 网宿科技股份有限公司 | Method and device for enhancing video image quality |
| CN113034412A (en) * | 2021-02-25 | 2021-06-25 | 北京达佳互联信息技术有限公司 | Video processing method and device |
| CN113034412B (en) * | 2021-02-25 | 2024-04-19 | 北京达佳互联信息技术有限公司 | Video processing method and device |
| CN114598902A (en) * | 2022-03-09 | 2022-06-07 | 安徽文香科技有限公司 | Video frame processing method and device and electronic equipment |
| CN114598902B (en) * | 2022-03-09 | 2023-12-22 | 安徽文香科技股份有限公司 | Video frame processing method and device and electronic equipment |
| CN115272142A (en) * | 2022-09-30 | 2022-11-01 | 南通市通州区华凯机械有限公司 | Scene image preprocessing method of immersive driving simulator |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106550244A (en) | The picture quality enhancement method and device of video image | |
| CN104067311B (en) | Digital makeup | |
| TW201723930A (en) | Generation method of color QR code | |
| CN110032989B (en) | Table document image classification method based on frame line characteristics and pixel distribution | |
| CN105741322A (en) | Region segmentation method of field of view on the basis of video feature layer fusion | |
| CN111667410B (en) | Image resolution improving method and device and electronic equipment | |
| CN106875358A (en) | Image enchancing method and image intensifier device based on Bayer format | |
| CN105096280A (en) | Method and device for processing image noise | |
| CN102333233A (en) | Stereo image quality objective evaluation method based on visual perception | |
| CN105095857B (en) | Human face data Enhancement Method based on key point perturbation technique | |
| CN104021523B (en) | A kind of method of the image super-resolution amplification based on marginal classification | |
| CN105989611A (en) | Blocking perception Hash tracking method with shadow removing | |
| CN104574328A (en) | Color image enhancement method based on histogram segmentation | |
| CN109584185A (en) | Image processing method | |
| CN104992419A (en) | Super pixel Gaussian filtering pre-processing method based on JND factor | |
| CN107610675A (en) | A kind of image processing method and device based on dynamic level | |
| CN105488475A (en) | Method for detecting human face in mobile phone | |
| CN108961188A (en) | A kind of image quality Enhancement Method, system and device | |
| CN113095992A (en) | Novel bar code screenshot steganography traceability combined algorithm | |
| Yang et al. | Full reference image quality assessment by considering intra-block structure and inter-block texture | |
| CN115131229A (en) | Image noise reduction and filtering data processing method and device and computer equipment | |
| CN110298835A (en) | A kind of leather surface damage testing method, system and relevant apparatus | |
| CN110378860A (en) | Method, apparatus, computer equipment and the storage medium of restored video | |
| CN110324617A (en) | Image processing method and device | |
| CN110399886B (en) | Screen image JND model construction method |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200526 Address after: 310052 room 508, floor 5, building 4, No. 699, Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Applicant after: Alibaba (China) Co.,Ltd. Address before: 510627 Guangdong city of Guangzhou province Whampoa Tianhe District Road No. 163 Xiping Yun Lu Yun Ping B radio 14 floor tower square Applicant before: GUANGZHOU UCWEB COMPUTER TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170329 |
|
| RJ01 | Rejection of invention patent application after publication |