CN109194923A - Video image processing apparatus, system and method based on non-uniformed resolution ratio - Google Patents
Video image processing apparatus, system and method based on non-uniformed resolution ratio Download PDFInfo
- Publication number
- CN109194923A CN109194923A CN201811212816.5A CN201811212816A CN109194923A CN 109194923 A CN109194923 A CN 109194923A CN 201811212816 A CN201811212816 A CN 201811212816A CN 109194923 A CN109194923 A CN 109194923A
- Authority
- CN
- China
- Prior art keywords
- region
- resolution
- video image
- mapping
- interest
- 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
- 238000012545 processing Methods 0.000 title claims abstract description 69
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000013507 mapping Methods 0.000 claims abstract description 89
- 230000005540 biological transmission Effects 0.000 claims abstract description 15
- 230000006835 compression Effects 0.000 claims abstract description 13
- 238000007906 compression Methods 0.000 claims abstract description 13
- 230000007704 transition Effects 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 5
- 239000012141 concentrate Substances 0.000 claims description 4
- 230000000149 penetrating effect Effects 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 17
- 238000004458 analytical method Methods 0.000 abstract description 5
- 230000000007 visual effect Effects 0.000 abstract description 4
- 230000001737 promoting effect Effects 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000003321 amplification Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000003199 nucleic acid amplification method Methods 0.000 description 3
- 239000000203 mixture Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000001235 sensitizing effect Effects 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000009365 direct transmission Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- 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]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- 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]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/4402—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving reformatting operations of video signals for household redistribution, storage or real-time display
- H04N21/440245—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving reformatting operations of video signals for household redistribution, storage or real-time display the reformatting operation being performed only on part of the stream, e.g. a region of the image or a time segment
-
- 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]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/472—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content
- H04N21/47205—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content for manipulating displayed content, e.g. interacting with MPEG-4 objects, editing locally
-
- 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]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/472—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content
- H04N21/4728—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content for selecting a Region Of Interest [ROI], e.g. for requesting a higher resolution version of a selected region
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Human Computer Interaction (AREA)
- Image Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses video image processing apparatus, system and methods based on non-uniformed resolution ratio, are related to technical field of image processing.The video image processing apparatus includes: memory, to storing data and instruction；Processor can determine at least one region-of-interest in monitor video image；And before Video coding or transmission, non-uniform resolution mapping processing is carried out to monitor video image and obtains small size video image, more resolution ratio wherein are distributed for the region-of-interest, while keeping rectangular shape and the size of every frame video image after mapping to encode compression.The present invention can existing bandwidth, hardware capabilities, storage cost of recording a video limitation under, take into account security area local video resolution ratio when guaranteeing monitoring visual field range and promoting video analysis, playback and preview.
Description
Technical field
The present invention relates to technical field of image processing, and in particular to a kind of image procossing based on non-uniformed resolution ratio
Technology.
Background technique
Video monitoring system in use, on the one hand user usually wishes that camera can monitor enough wide visual field models
It encloses, monitored picture can show more region, on the other hand want to see local details again, this just needs to monitor
System can support image resolution ratio as big as possible.But the image resolution limitations that can actually use of video monitoring system in
The influence of video system transmission bandwidth, hardware processing capability, video recording storage cost etc., tends not to using monitor camera most
Big resolution ratio, but lesser resolution ratio can only be used.Therefore, if limited according to user demand and system performance, adaptively
Control video resolution, it will help to promote the performance of monitoring system.
The method of a variety of resolution ratio self adaptive controls is proposed in existing monitoring technology research.Such as: in order to adapt to not
Same transmission bandwidth and hardware capabilities, some monitor video systems use a variety of resolution ratio code stream (primary keys of code storage simultaneously
Stream and time code stream) method, select the encoding code stream of a certain resolution ratio to be played back according to hardware playback capability.This method is suitable
There are the feelings of certain surplus when comparing independent primary key stream process for video recording system hardware processing capability and video transmission bandwidth
Condition.In some monitoring systems, in order to improve the clarity in the regions such as face, using the algorithm analysis identification people of artificial intelligence
The area-of-interests such as face, control video encoder distribute these regions more code rate, to reduce these regions because of video
Mass loss caused by compression.But if original input picture resolution ratio is not high, the effect of this method is with regard to limited.Some
Monitoring applied compression transmits the fish eye lens picture that do not correct, to keep the high-resolution at picture center, but since picture is turned round
Song, video compression efficiency is impacted, and the flake video for directly displaying distortion also will affect user and identify that object and judgement distance close
System, and monitor sensitizing range and be not necessarily located in picture center, it is more using limiting.
Currently, in monitoring application, (virtual warning lines or virtual alert are arranged in user to boundary defence on video background picture
Guard against area) it is a kind of common application scenarios.It is a kind of in application, on the one hand user needs picture field range vast, can include
The full scope in the region that one or more needs to guard against, and there are enough viewing areas outside warning region to observe
The process of people or object disengaging security area；Still further aspect, user are different the clarity demand of monitored picture different zones,
Other regions are much higher than to the interest of local warning region.Especially when event occurs for warning region, user has strong
Demand carries out preview after amplifying to warning region picture or playback is watched, but if the original resolution of monitor video is lower,
Even if image amplifies, details still can not be seen clearly.
Summary of the invention
It is an object of the invention to: it has overcome the deficiencies of the prior art and provide a kind of based on non-uniformed resolution ratio
Video image processing apparatus, system and method.The monitor video image procossing of non-uniformed resolution ratio proposed by the present invention is set
It is standby, can existing bandwidth, hardware capabilities, storage cost of recording a video limitation under, take into account and guarantee monitoring visual field range and promote view
Security area local video resolution ratio when frequency analysis, playback and preview.
To realize above-mentioned target, the present invention provides the following technical scheme that
The present invention provides a kind of video image processing apparatus based on non-uniformed resolution ratio comprising:
Memory, to storing data and instruction；
Processor is used for when executing described instruction,
Obtain monitor video image；
Determine at least one region-of-interest in monitor video image；
And before Video coding or transmission, non-uniform resolution mapping processing is carried out to monitor video image and obtains small ruler
Very little video image wherein distributing more resolution ratio for the region-of-interest, while keeping every frame video image after mapping
Rectangular shape and size are compressed to encode.
Further, the processor is used in video playback or preview,
It carries out resolution ratio reflection and penetrates processing, so that the reflection of small size video image is emitted back towards recovery figure when large scale shows image
As proportionate relationship, high-resolution image is exported in the region-of-interest partial enlargement.
Further, the processor includes region-of-interest setup module, is used for,
At least one region-of-interest is generated on the video images based on being manually entered；
And/or the scene characteristic based on capture automatically generates at least one region-of-interest on the video images；
And/or the region-of-interest is edited to reconfigure the region-of-interest of video image, the editor includes that adjustment is closed
The range in region is infused, new region-of-interest is added and/or deletes existing region-of-interest.
Further, the processor is also used to when determining at least one region-of-interest,
The local resolution that region-of-interest is arranged promotes multiple m；
Resolution ratio transitional region is set around region-of-interest, point of the pixel except the region-of-interest and transitional region
Resolution keeps the low resolution of original system setting to form low resolution region, the shadow that low resolution region is not mapped by resolution ratio
It rings；
The local resolution promotes multiple m and is less than the scale ratio of transitional region and corresponding region-of-interest in all directions most
Small value.
Further, the processor is used for when generating small size video image,
The different zones of video image are reduced using different resolution ratio mapping modes, the region-of-interest conduct
High-resolution areas is scaled using diminution ratio k, and the low resolution region is scaled using diminution ratio L, k=L/m.
Further, the mapping mode of the transitional region, using transitting probability method or multinomial projection's transform method.
The present invention also provides a kind of video image processing system based on non-uniformed resolution ratio, the system packet
It includes:
Monitor camera, to obtain high-resolution raw video image；
Image processing equipment, it is described to carry out non-uniformed resolution ratio mapping processing to aforementioned raw video image
Image processing equipment is equipment of any of claims 1-6；
Display is monitored, to show through described image processing equipment treated video image.
The present invention also provides a kind of method of video image processing based on non-uniformed resolution ratio, including walk as follows
It is rapid:
Obtain monitor video image；
Determine at least one region-of-interest in monitor video image；
Before Video coding or transmission, non-uniform resolution mapping processing is carried out to original monitor video image and obtains small ruler
Very little video image wherein distributing more resolution ratio for the region-of-interest, and keeps the square of every frame video image after mapping
Shape shapes and sizes；
To non-uniform resolution mapping, treated that video image carries out coding compression.
Further, non-uniform resolution mapping processing is carried out to original monitor video image and obtains small size video image
Step includes,
The local resolution that region-of-interest is arranged promotes multiple m, and resolution ratio transitional region, institute are arranged around region-of-interest
The resolution ratio for stating the pixel except region-of-interest and transitional region keeps the low resolution of original system setting to form low resolution area
Domain, the local resolution promote multiple m and are less than scale ratio minimum value of the transitional region with corresponding region-of-interest in all directions；
The different zones of video image are reduced using different resolution ratio mapping modes, the region-of-interest conduct
High-resolution areas is scaled using diminution ratio k, and the low resolution region is scaled using diminution ratio L, k=L/m；The mistake
The mapping mode in region is crossed using transitting probability method or multinomial projection's transform method.
The present invention also provides a kind of non-uniform resolution mapping method based on perspective transform, includes the following steps:
Step 1, high-resolution areas and transitional region boundary are set on the original image；The high-resolution areas and point
Resolution transitional region is the shape of arbitrary polygon, and two polygons have identical vertex quantity n, records two sides n on image
The corresponding relationship of shape apex coordinate and each side end point；
Step 2, it reorders clockwise to zone boundary polygon vertex；First from the vertex of transitional region boundary polygon
The middle highest vertex A of selection vertical direction position, when having multiple height identical vertex selection wherein horizontal position most
The vertex on the left side, according to the connected relation on side, is in the direction of the clock transitional region boundary polygon and high score from A point
It sorts on each vertex of resolution zone boundary polygon；
Step 3, the vertex of transitional region and high-resolution areas is matched two-by-two；
A. to the vertex A on transitional region boundary, calculate whether each vertex in high-resolution areas boundary meets with it and can match
Condition；
B. it is concentrated from the pairing candidate point of A point, selects a high-resolution areas vertex and A point trial pair, detect transition
Whether the examination match point of each vertex correspondence in region, which all meets, can match condition；
Trial is concentrated from the pairing candidate point of A point and is removed to failure when in the presence of the trial opposite vertexes for being unsatisfactory for pairing condition
This high-resolution areas vertex；Otherwise, successful matching is tried, regard the distance summation of two each match points of polygon as this point
Pairing cost value；
C. b step is recycled, the pairing candidate point until completing A point concentrates all candidate points and the trial pair of A；
D. compare the pairing cost value with A point trial to successful each point, select the smallest candidate point of cost as final
Match point, and complete high-resolution areas vertex in the direction of the clock and each vertex of transitional region is matched；
Step 4, each apex coordinate on target mapping graph is calculated；
Step 5, transitional region is divided into several Mapping quadrilaterals；
Step 6, quadrangle mapping and anti-mapping matrix are solved；According to perspective transform formula, to every group of corresponding mapping four
Side shape solves unique mapping and anti-mapping matrix.
Further, further include following steps:
Mapping is carried out to high-definition picture and reduces processing acquisition small size video image, and/or in video playback or in advance
When looking at, reflection is carried out to small size video image and penetrates enhanced processing；
In the treatment process, when target pixel points are located at high-resolution areas or low resolution region, according to correspondence
Zoom factor carries out mapping processing；When target pixel points are located at resolution ratio transitional region, judge that it is located at which perspective transform is reflected
It penetrates in quadrangle, then carries out mapping processing using corresponding perspective transform relationship.
The present invention due to using the technology described above, compared with prior art, as an example, has the following advantages that and accumulates
Pole effect:
1) present invention can be according to user to the different demands of picture different zones resolution ratio, on the video image of diminution
Transmit non-uniform resolution information.So that even if because the reasons such as bandwidth, hardware performance, storage cost are by original camera image
Minification transmission and coding, are still able to maintain the high resolution information of the security area of user setting, help to expand monitoring picture
The angular field of view that face allows promotes detecting/region intrusion detecting scheduling algorithm accuracy of crossing the border, and convenient for amplifying when user's playback
Observe object detail.
2) monitor video image is divided into high-resolution, low resolution and resolution ratio transition according to local features by the present invention
The method in the three kinds of regions in region, because high-resolution and low resolution area image are not distorted, it is possible to reduce frequently horizontal
The decline of Video coding compression ratio caused by the scalloping of moving region helps to reduce monitor video storage cost.This is than existing
Direct transmission fisheye image, with guarantee the high-resolution of picture centre be more suitable for coding compression.
3) mapping method of the further disclosed resolution ratio transitional region based on perspective transform of the present invention, has adaptation each
Kind of transitional region shape, the characteristics of change resolution is gentle, cost is low, easy realization, and advantageously reduce mapping scalloping
Degree, reduce scalloping and restore bring mass loss.
Detailed description of the invention
Fig. 1 is the modular structure schematic diagram of monitor video system commonly used in the prior art.
Fig. 2 is the module knot of the video image processing system provided in an embodiment of the present invention based on non-uniformed resolution ratio
Structure schematic diagram.
Fig. 3 is that the big figure of high-resolution provided in an embodiment of the present invention carries out non-uniformed mapping acquisition small-sized image
Display example figure.
Fig. 4 is the pairing candidate point exemplary diagram of vertex A provided in an embodiment of the present invention.
Fig. 5 is divided into several corresponding perspectives with mapping image transition zone for original image provided in an embodiment of the present invention
Convert the exemplary diagram of quadrangle.
Fig. 6 is the process of the method for video image processing provided in an embodiment of the present invention based on non-uniformed resolution ratio
Figure.
Fig. 7 is the flow chart of the non-uniform resolution mapping method provided in an embodiment of the present invention based on perspective transform.
Description of symbols:
Monitor camera 110, image processing equipment 120 monitor display 130.
Specific embodiment
Below in conjunction with the drawings and specific embodiments to the video image disclosed by the invention based on non-uniformed resolution ratio
Processing equipment, system and method are described in further detail.It should be noted that following embodiments described in technical characteristic or
The combination of person's technical characteristic be not construed as it is isolated, they can by be combined with each other to reach superior technique effect
Fruit.In the drawings of the following embodiments, the identical label that each attached drawing occurs represents identical feature or component, can be applied to
In different embodiments.Therefore, it once being defined in a certain Xiang Yi attached drawing, then does not need to carry out it in subsequent attached drawing
It further discusses.
It should be noted that structure, ratio, size etc. depicted in this specification institute attached drawing, only to cooperate explanation
The revealed content of book is not limited to invent enforceable restriction item so that those skilled in the art understands and reads
Part, the modification of any structure, the change of proportionate relationship or the adjustment of size are not influencing the effect of invention can be generated and institute's energy
Under the purpose reached, it should all fall in the range of the revealed technology contents of invention can cover.Preferred implementation side of the invention
The range of formula includes other realization, wherein described or discussion the sequence can not be pressed, including is pressed according to related function
Basic mode simultaneously or in the opposite order, to execute function, this should be by the skill of the embodiment of the present invention technical field
Art personnel are understood.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as authorizing part of specification.Institute that is shown here and discussing
Have in example, any occurrence should be construed as merely illustratively, not as limitation.Therefore, exemplary embodiment
Other examples can have different values.
Embodiment
Video monitoring system usually can be by video camera, video analysis and processing, transmission of video, video record, video
Several module compositions such as playback and display.For analog video camera, monitoring system transmission is analog video, and video is connecing
Receiving end is analyzed and processed, code storage, plays back display；For digital video camcorder, monitoring system is just carried out in transmitting terminal
Analysis processing and coding compression only need to do storage and playback display in receiving end.
In monitor video application, (virtual warning lines or virtual alert are arranged in user to boundary defence on video background picture
Guard against area) it is a kind of common application scenarios.It is this kind of in application, on the one hand user needs picture field range vast, can wrap
Full scope containing the region that one or more needs are guarded against, and there are enough viewing areas outside warning region to see
Examine the process of people or object disengaging security area；Still further aspect, user are different to the clarity demand of monitored picture different zones
, other regions are much higher than to the interest of local warning region.Especially when event occurs for warning region, user has strongly
Demand to warning region picture amplification after carry out preview or playback watch, need to see clearly the details such as face and article.
That is, the virtual security area is the region-of-interest of user, it is expected to configure high-resolution.But in fixed resolution
Under rate, when picture field range increases, the resolution ratio of each regional area will be reduced.
It is shown in Figure 1, it is monitor video system commonly used in the prior art, the high-resolution that monitor camera obtains is big
Figure obtains the small figure of low resolution after diminution is handled, and then carries out Video coding compression and storage to small figure；In video playback
Or when preview, video is decoded, the small figure of low resolution is exported to user, user can according to need to small figure into putting
Greatly, monitor display output at this time is the big figure of low resolution, and the clarity of image is difficult to ensure.
In order to guarantee the visual field, the resolution ratio of image can only be sacrificed.When local resolution is lower, detecting/region of crossing the border
The accuracy of intrusion detecting scheduling algorithm will receive influence, even and if user image is put in playback of monitoring videos video
Big operation, still can not see details clearly.Also, when sensitizing range resolution ratio is lower, detecting of crossing the border/region intrusion detecting scheduling algorithm
Accuracy also will receive influence.
It is shown in Figure 2, it is a kind of video image processing system based on non-uniformed resolution ratio provided by the invention.
The video image processing system includes the monitor camera 110 of communication connection, image processing equipment 120 and monitoring
Display 130.
The monitor camera 110, to obtain high-resolution raw video image.The monitor camera 110 can
To be digital video camcorder, it is also possible to analog video camera.
Described image processing equipment 120, to be carried out at the mapping of non-uniformed resolution ratio to aforementioned raw video image
Reason.
The monitoring display 130, to show through described image processing equipment treated video image.As an example
Rather than limit, the monitoring display can be liquid crystal display.
The network of the data transmission of each section can be Intranet, internet or one or more any kind of in system
The combination of a wired or wireless network or one or more wired or wireless networks.
Described image processing equipment 120 specifically may be used based on the video image processing apparatus of non-uniformed resolution ratio
To include memory and processor.
Memory, to storing data and instruction.
Processor is used for when executing described instruction,
Obtain monitor video image；
Determine at least one region-of-interest in monitor video image；
And before Video coding or transmission, non-uniform resolution mapping processing is carried out to monitor video image and obtains small ruler
Very little video image wherein distributing more resolution ratio for the region-of-interest, while keeping every frame video image after mapping
Rectangular shape and size are compressed to encode.
In the present embodiment, described image processing equipment 120 may include the memory for storing one or more instructions, and
For executing the processor of one or more of instructions, upon being performed, these instructions can configure the processor to mention
For function described herein.Certainly, described image processing equipment can also include other groups usually found in calculating equipment
Part, such as inputting information to processing equipment or from one or more input output assemblies of processing equipment output information, than
Such as video camera, keyboard, mouse, network adapter.
With continued reference to shown in Fig. 2, the processor is also used in video playback or preview: being carried out resolution ratio reflection and is penetrated place
Reason, the small size video image reflection that decoding is obtained, which is emitted back towards when large scale shows image, restores image scaled relationship,
And high-resolution image is exported in the region-of-interest partial enlargement.
In the present embodiment, the processor may include having region-of-interest setup module.The region-of-interest setup module
For: at least one region-of-interest is generated on the video images based on being manually entered；And/or the scene characteristic based on capture exists
At least one region-of-interest is automatically generated on video image；And/or the region-of-interest is edited to reconfigure video image
Region-of-interest, the editor include the range of adjustment region-of-interest, the new region-of-interest of addition and/or delete and have concern area
Domain.
In this way, region-of-interest (for the desired region with higher resolution of user) of the invention can be in many ways
It determines.For example, being arranged one or more region-of-interests based on being manually entered for user.In another embodiment, area is paid close attention to
Domain can be determined automatically, be preferably based on the known of captured scene or the feature determined (for example, fence on road,
Cut-off rule etc.).
Preferably, the region-of-interest setup module supports user as needed, and fixed pass is arranged on monitored picture
It infuses region (high-resolution area) and local resolution promotes multiple m, and set impacted resolution ratio mistake around region-of-interest
Cross region.
The local resolution, which promotes multiple m, need to be less than scale ratio of the transitional region with corresponding region-of-interest in all directions
Minimum value.The resolution ratio of pixel except each region-of-interest and transitional region keeps the low resolution of original system setting ---
It is used as low resolution region, is not influenced by resolution ratio mapping, the more outer frequent area that object horizontal movement occurs of area will be paid close attention to
Domain is left low resolution area, can be with the coding compression efficiency of lift map rear video.
Shown in Figure 3, for the example of region-of-interest setting, the high-resolution areas as region-of-interest, which is located at, divides
The middle position of resolution transition region.Certainly, above-mentioned division mode by way of example and not limitation, the high score as region-of-interest
Resolution region can be located at the middle position of resolution ratio transition region, can also be located at the edge or picture of resolution ratio transition region
Any other position.Compared with the existing technology, the setting of the region-of-interest of the invention is unrestricted.
When processor processing, resolution ratio mapping block can use division region different on the big figure of original video different
Resolution ratio mapping mode reduced:
The high-resolution areas is zoomed in and out using lower image down ratio k, to the low resolution region
It is scaled using image down ratio L, the k=L/m.
Then, the resolution ratio transitional region of original image is mapped to by remaining sky on the small figure of target by the mapping mode of design
White region.The mapping mode of the transitional region can be carried out using the methods of perspective transform, multinomial projection's transformation.
It is the small figure of non-uniform resolution that the big figure of high-resolution obtains after mapping shown in the following figure referring to Fig. 3.Mapping
There is no variations for the shapes and sizes for the small figure completed, therefore can carry out compression volume using traditional coding and decoding video module
Code and decoding.
The image content for mapping the small figure completed can be distorted deformation, observe for the ease of user, and the processor is set
It is equipped with anti-mapping block, the anti-mapping block is used for according to different resolution region division and corresponding resolution ratio mapping side
Method replys image scaled relationship while amplifying small figure, eliminate picture deformation.
In this way, the high-resolution areas of original image setting, is handled by system above, high score still can be kept in display
Resolution, and then convenient for observing the details of image when user's amplification.
It is shown in Figure 4, preferably, carrying out the typical case of the method for image procossing using above-mentioned video image processing system
Steps are as follows:
S110 obtains monitor video image.
S120 determines at least one region-of-interest in monitor video image.
S130 carries out non-uniform resolution mapping processing to original monitor video image and obtains before Video coding or transmission
Small size video image is obtained, wherein distributing more resolution ratio for the region-of-interest, and keeps every frame video figure after mapping
The rectangular shape and size of picture.
S140, to non-uniform resolution mapping, treated that video image carries out coding compression.
In the step S130, non-uniform resolution mapping processing is carried out to original monitor video image and obtains small size view
The step of frequency image, specifically can be such that
The local resolution that the region-of-interest is arranged promotes multiple m, and resolution ratio transition region is arranged around region-of-interest
Domain, the resolution ratio of the pixel except the region-of-interest and transitional region keep the low resolution of original system setting to form low resolution
Rate region, the local resolution promote multiple m and are less than the scale ratio of transitional region and corresponding region-of-interest in all directions most
Small value；
The different zones of video image are reduced using different resolution ratio mapping modes, the region-of-interest conduct
High-resolution areas is scaled using diminution ratio k, and the low resolution region is scaled using diminution ratio L, k=L/m；The mistake
The mapping mode in region is crossed using transitting probability method or multinomial projection's transform method.
Mapping for resolution ratio transitional region, the non-uniform resolution mapping based on perspective transform that this paper presents a kind of
Method, shown in Figure 5, described method includes following steps:
1) high-resolution areas and transitional region boundary are set on the original image.
Preferably, high-resolution areas that user sets on the original image and resolution ratio transitional region are limited to be any more
The shape of side shape, it can be convex polygon or concave polygon that two polygons, which must have the side identical vertex quantity n, n shape,.Remember respectively
Record the corresponding relationship of two side n shape apex coordinates and each side end point on original image.
2) zone boundary polygon vertex reorders clockwise.
The highest vertex A of vertical direction position is selected first from the vertex of transitional region boundary polygon, if having
The identical vertex of multiple height then select wherein horizontal position on leftmost vertex.Then, according to the area polygonal of record
Side and endpoint corresponding relationship can obtain two sides being connected with vertex A and corresponding vertex.Select the positive water from A point
Flat axis is rotated in the clockwise direction the first side encountered, and is from A point along the direction on this edge traversal each vertex of polygon
Clockwise.It is in the direction of the clock each vertex of transitional region boundary polygon according to the connected relation on side from A point
It is ranked up.It is denoted as A, B, C, D etc. respectively.
Same treatment is carried out to high-resolution area boundary polygon.
3) transitional region and high-resolution areas vertex are matched two-by-two.
A. to the vertex A on transitional region boundary, calculate whether each vertex in high-resolution areas boundary meets with it and can match
Condition.
Shown in Figure 6, in this implementation, the vertex A's matches condition are as follows: for the high-resolution areas top matched
Point and the line of A must be positioned within the angular range on two sides of A, and with each of high-resolution areas boundary side
It is non-intersecting.
The high-resolution areas vertex composition A point for meeting condition matches candidate point set.If matching candidate point set is sky,
Region setup failed, return step 1) prompt user resets the shape of transitional region and high-resolution areas.
B. it is concentrated from the pairing candidate point of A point, selects a high-resolution areas vertex and A point trial pair: by two sides n
The remaining each vertex of shape is according to successively trial pair, the examination match point of detection each vertex correspondence of transitional region are two-by-two clockwise
No all satisfactions can match condition.
If there is the trial opposite vertexes for being unsatisfactory for pairing condition, then trial concentrates failure from the pairing candidate point of A point
Remove this high-resolution area vertex；Otherwise, successful matching is tried, and regard the summation of the distance of two each match points of polygon as this
The pairing cost value of a bit.
C. above-mentioned b step is recycled, the pairing candidate point until completing A point concentrates all candidate points and the trial pair of A.If
Candidate point is concentrated there is no trial is met to successful candidate point, then region setup failed, return step 1) prompt user to reset
The shape of transitional region and high-resolution areas.
D. compare the pairing cost value with A point trial to successful each point, the smallest candidate point of cost value is selected, as most
Whole match point, and high-resolution areas vertex and each vertex pairing of transitional region are completed in the direction of the clock.It can with opposite vertexes
It is shown in Figure 7 labeled as (A0, a0) (B0, b0) (C0, c0) (D0, d0) etc..
4) each apex coordinate on target mapping graph is calculated.
Ratio L is reduced according to picture size and local resolution promotes multiple m, calculates each region vertex in the small figure of mapping
On coordinate.It can be labeled as (A1, a1) (B1, b1) (C1, c1) (D1, d1) etc..If mapping the high-resolution area on small figure
Vertex (a1, b1, c1, d1 etc.) is located at except transitional region area boundary, then the local resolution of user setting promotes multiple m mistake
Greatly, local resolution need to be reduced and promote multiple m value, repeat step 4), reach reasonable model until local resolution promotes multiple m value
It encloses.
5) transitional region is divided into several Mapping quadrilaterals.
Judge two endpoints and their corresponding match points on original image transitional region any bar side, if be in one
On straight line.If not point-blank, they form an original image transitional region quadrangle to this four points, such as
A0B0b0a0.This four points map the corresponding points on small figure in target and form a corresponding target figure transitional region quadrangle,
It is shown in Figure 7 such as A1B1b1a1.In this way, original image transitional region and target figure transitional region can be separately disassembled into 2~n
Mapping quadrilateral, and correspond.
6) quadrangle mapping and anti-mapping matrix are solved.
According to perspective transform formula, to every group of corresponding Mapping quadrilateral, unique mapping can be solved and square is penetrated in reflection
Battle array.
Further, the method also includes following steps:
7) (diminution) processing is mapped to original high-resolution image, obtains small size video image.
First judge which region is target pixel points be in, if pixel is located at high-resolution areas or low resolution region,
Then processing is zoomed in and out according to corresponding zoom factor；If object pixel is located at resolution ratio transitional region, judge which it is located at
In a perspective transform Mapping quadrilateral, then the corresponding perspective transform relationship of application carries out pixel-map.
8) in video playback or preview, reflection is carried out to small size video image and penetrates (amplification) processing, with step 7) side
Method is identical.
In the treatment process, when target pixel points are located at high-resolution areas or low resolution region, according to correspondence
Zoom factor carries out mapping processing；When target pixel points are located at resolution ratio transitional region, judge that it is located at which perspective transform is reflected
It penetrates in quadrangle, then carries out mapping processing using corresponding perspective transform relationship.
Non-uniform resolution Monitor Equipment proposed by the present invention can be arranged, video camera according to the security area of user
High Resolution and Large Size video image by non-uniform resolution map, be mapped on small size video image transmit, encode,
Storage.Higher resolution ratio is remained in small size video for security area.In video tour or playback, small size video figure
Image is shown as reflection is emitted back towards large scale, obtains high-resolution in display image security area.Further, by monitor video image point
For three kinds of high-resolution, low resolution and resolution ratio transitional region regions, reflected using different resolution ratio mapping methods
It penetrates, and synthesizes small size rectangular target mapping graph, be convenient for coding transmission.And optimize the mapping method of resolution ratio transitional region.
It will be apparent to one skilled in the art that determining that region-of-interest can also use other conventional methods, for example it is based on catching
The view of multiple video cameras of scene is obtained to determine the overlay area of overlapping, and the region of the overlapping is determined as region-of-interest.
Although above example, which is shown, has only gone out 1 region, but it is to be understood that there may be the region-of-interests with 2 or more.
Aforementioned processor can be any kind of processor, such as general Central Processing Unit (" CPU ") or such as embedding
Enter the special microprocessor or digital signal processor (" DSP ") of the controller etc that declines.
Memory can be suitable for storing and accessing any kind of memory of electronic information or their combination, all
Such as temporary random access memory (RAM) or non-transitory memory, such as read-only memory (ROM), disk drive memory
Device, archival memory, CD drive memory, optical memory etc..
Memory may include data and instruction, and when being executed by a processor, described instruction can configure or make equipment
Execute or implement function described above or aspect (for example, processing one or more steps).In addition, equipment can also include logical
The other components often found in computing systems such as store in memory and by the operating system of processor execution, queue
Manager, device driver, database device or one or more network protocols etc..
In the above description, the disclosure is not intended to for its own to be limited to these aspects.But at this
Within the scope of the protection of goal of disclosure, each component can selectively and operatively be merged with arbitrary number.In addition,
As the term of " comprising ", " including " and " having " should default it is being interpreted as including property or open, rather than it is exclusive
Property or closure, unless it is explicitly defined as opposite meaning.All technologies, science and technology or otherwise term all meet
Meaning understood by one of ordinary skill in the art, unless it is defined as opposite meaning.The public term found in dictionary is answered
It is not idealized very much or is impractically explained very much when under the background in the relevant technologies document, unless present disclosure is clearly limited
Determine at such.Any change, the modification that the those of ordinary skill in field of the present invention does according to the disclosure above content, belong to right
The protection scope of claim.
Claims (11)
1. a kind of video image processing apparatus based on non-uniformed resolution ratio, characterized by comprising:
Memory, to storing data and instruction；
Processor is used for when executing described instruction,
Obtain monitor video image；
Determine at least one region-of-interest in monitor video image；
And before Video coding or transmission, non-uniform resolution mapping processing is carried out to monitor video image and obtains small size view
Frequency image wherein distributing more resolution ratio for the region-of-interest, while keeping the rectangle of every frame video image after mapping
Shapes and sizes are compressed to encode.
2. equipment according to claim 1, it is characterised in that: the processor is used in video playback or preview,
It carries out resolution ratio reflection and penetrates processing, so that the reflection of small size video image, which is emitted back towards when large scale shows image, restores image ratio
Example relationship, exports high-resolution image in the region-of-interest partial enlargement.
3. equipment according to claim 1 or 2, it is characterised in that: the processor includes region-of-interest setup module,
For,
At least one region-of-interest is generated on the video images based on being manually entered；
And/or the scene characteristic based on capture automatically generates at least one region-of-interest on the video images；
And/or the region-of-interest is edited to reconfigure the region-of-interest of video image, the editor includes adjustment concern area
The range in domain adds new region-of-interest and/or deletes existing region-of-interest.
4. equipment according to claim 1 or 2, it is characterised in that: the processor is determining at least one region-of-interest
When be also used to,
The local resolution that region-of-interest is arranged promotes multiple m；
Resolution ratio transitional region, the resolution ratio of the pixel except the region-of-interest and transitional region are set around region-of-interest
The low resolution of original system setting is kept to form low resolution region, low resolution region is not influenced by resolution ratio mapping；
The local resolution promotes multiple m and is less than the scale ratio minimum of transitional region and corresponding region-of-interest in all directions
Value.
5. equipment according to claim 4, it is characterised in that: the processor is used when generating small size video image
In,
The different zones of video image are reduced using different resolution ratio mapping modes, the region-of-interest is as high score
Resolution region is scaled using diminution ratio k, and the low resolution region is scaled using diminution ratio L, k=L/m.
6. equipment according to claim 5, it is characterised in that: the mapping mode of the transitional region, using transitting probability
Method or multinomial projection's transform method.
7. a kind of video image processing system based on non-uniformed resolution ratio, characterized by comprising:
Monitor camera, to obtain high-resolution raw video image；
Image processing equipment, to carry out non-uniformed resolution ratio mapping processing, described image to aforementioned raw video image
Processing equipment is equipment of any of claims 1-6；
Display is monitored, to show through described image processing equipment treated video image.
8. a kind of method of video image processing based on non-uniformed resolution ratio, it is characterised in that include the following steps:
Obtain monitor video image；
Determine at least one region-of-interest in monitor video image；
Before Video coding or transmission, non-uniform resolution mapping processing is carried out to original monitor video image and obtains small size view
Frequency image wherein distributing more resolution ratio for the region-of-interest, and keeps the rectangle shape of every frame video image after mapping
Shape and size；
To non-uniform resolution mapping, treated that video image carries out coding compression.
9. according to the method described in claim 8, being reflected it is characterized by: carrying out non-uniform resolution to original monitor video image
Penetrating the step of processing obtains small size video image includes,
The local resolution that region-of-interest is arranged promotes multiple m, and resolution ratio transitional region, the pass are arranged around region-of-interest
The resolution ratio for infusing the pixel except region and transitional region keeps the low resolution of original system setting to form low resolution region, institute
It states local resolution and promotes multiple m less than scale ratio minimum value of the transitional region with corresponding region-of-interest in all directions；
The different zones of video image are reduced using different resolution ratio mapping modes, the region-of-interest is as high score
Resolution region is scaled using diminution ratio k, and the low resolution region is scaled using diminution ratio L, k=L/m；The transition region
The mapping mode in domain uses transitting probability method or multinomial projection's transform method.
10. a kind of non-uniform resolution mapping method based on perspective transform, it is characterised in that include the following steps:
Step 1, high-resolution areas and transitional region boundary are set on the original image；The high-resolution areas and resolution ratio
Transitional region is the shape of arbitrary polygon, and two polygons have identical vertex quantity n, records two sides n shape top on image
The corresponding relationship of point coordinate and each side end point；
Step 2, it reorders clockwise to zone boundary polygon vertex；It is selected from the vertex of transitional region boundary polygon first
Select the highest vertex A of vertical direction position, when having multiple height identical vertex selection wherein horizontal position in Far Left
Vertex from A point, according to the connected relation on side, be in the direction of the clock transitional region boundary polygon and high-resolution
It sorts on each vertex of zone boundary polygon；
Step 3, the vertex of transitional region and high-resolution areas is matched two-by-two；
A. to the vertex A on transitional region boundary, whether the calculating each vertex in high-resolution areas boundary meets with it can match condition；
B. it is concentrated from the pairing candidate point of A point, selects a high-resolution areas vertex and A point trial pair, detect transitional region
Whether the examination match point of each vertex correspondence, which all meets, can match condition；
Trial is concentrated from the pairing candidate point of A point to failure and removes this when in the presence of the trial opposite vertexes for being unsatisfactory for pairing condition
High-resolution areas vertex；Otherwise, successful matching is tried, the distance of two each match points of polygon is summed matching as this point
To cost value；
C. b step is recycled, the pairing candidate point until completing A point concentrates all candidate points and the trial pair of A；
D. compare the pairing cost value with A point trial to successful each point, the smallest candidate point of cost is selected to match as final
To point, and high-resolution areas vertex and each vertex pairing of transitional region are completed in the direction of the clock；
Step 4, each apex coordinate on target mapping graph is calculated；
Step 5, transitional region is divided into several Mapping quadrilaterals；
Step 6, quadrangle mapping and anti-mapping matrix are solved；According to perspective transform formula, to every group of corresponding Mapping quadrilateral
Solve unique mapping and anti-mapping matrix.
11. according to the method described in claim 10, characterized by further comprising following steps:
Mapping is carried out to high-definition picture and reduces processing acquisition small size video image, and/or in video playback or preview,
Reflection is carried out to small size video image and penetrates enhanced processing；
In the treatment process, when target pixel points are located at high-resolution areas or low resolution region, scaled according to corresponding
Coefficient carries out mapping processing；When target pixel points are located at resolution ratio transitional region, judge which perspective transform mapping four it is located at
In the shape of side, mapping processing then is carried out using corresponding perspective transform relationship.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811212816.5A CN109194923B (en) | 2018-10-18 | 2018-10-18 | Video image processing device, system and method based on local non-uniform resolution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811212816.5A CN109194923B (en) | 2018-10-18 | 2018-10-18 | Video image processing device, system and method based on local non-uniform resolution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109194923A true CN109194923A (en) | 2019-01-11 |
| CN109194923B CN109194923B (en) | 2023-08-29 |
Family
ID=64945442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811212816.5A Active CN109194923B (en) | 2018-10-18 | 2018-10-18 | Video image processing device, system and method based on local non-uniform resolution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109194923B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110572656A (en) * | 2019-09-19 | 2019-12-13 | 北京视博云科技有限公司 | coding method, image processing method, device and system |
| CN110796997A (en) * | 2019-11-14 | 2020-02-14 | 京东方科技集团股份有限公司 | Method and device for realizing non-uniform resolution display |
| WO2020173414A1 (en) * | 2019-02-25 | 2020-09-03 | 昀光微电子（上海）有限公司 | Human vision characteristic-based near-eye display method and device |
| CN111698453A (en) * | 2019-03-11 | 2020-09-22 | 杭州海康威视系统技术有限公司 | Video processing method and device |
| CN111988661A (en) * | 2019-05-24 | 2020-11-24 | 米利雅得广告公开股份有限公司 | Incorporating visual objects into video material |
| WO2021057684A1 (en) * | 2019-09-27 | 2021-04-01 | 腾讯科技（深圳）有限公司 | Video decoding method and apparatus, video encoding method and apparatus, storage medium, and electronic apparatus |
| WO2021073336A1 (en) * | 2019-10-18 | 2021-04-22 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | A system and method for creating real-time video |
| CN112884803A (en) * | 2020-08-18 | 2021-06-01 | 眸芯科技（上海）有限公司 | Real-time intelligent monitoring target detection method and device based on DSP |
| CN113518249A (en) * | 2020-04-10 | 2021-10-19 | 华为技术有限公司 | Far-end image processing method and device |
| CN113592712A (en) * | 2021-07-30 | 2021-11-02 | 北京百度网讯科技有限公司 | Image processing method, device, equipment, storage medium and cloud VR system |
| CN113630622A (en) * | 2021-06-18 | 2021-11-09 | 中图云创智能科技(北京)有限公司 | Panoramic video image processing method, server, target device, apparatus and system |
| CN114274877A (en) * | 2021-11-29 | 2022-04-05 | 惠州华阳通用智慧车载系统开发有限公司 | Rearview mirror image display control method |
| CN114520873A (en) * | 2021-12-31 | 2022-05-20 | 广州文远知行科技有限公司 | Sweeper video acceleration method and system |
| CN116761019A (en) * | 2023-08-24 | 2023-09-15 | 瀚博半导体(上海)有限公司 | Video processing method, system, computer device and computer readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120275718A1 (en) * | 2011-04-27 | 2012-11-01 | Tetsuya Takamori | Image compression apparatus, image expansion apparatus, and methods and programs thereof |
| CN105635624A (en) * | 2014-10-27 | 2016-06-01 | 华为技术有限公司 | Video image processing method, device and system |
| CN106791854A (en) * | 2016-11-22 | 2017-05-31 | 北京疯景科技有限公司 | Image Coding, coding/decoding method and device |
| CN108475420A (en) * | 2015-11-16 | 2018-08-31 | 阿尔卡特朗讯美国公司 | Multiresolution compressed sensing image procossing |
-
2018
- 2018-10-18 CN CN201811212816.5A patent/CN109194923B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120275718A1 (en) * | 2011-04-27 | 2012-11-01 | Tetsuya Takamori | Image compression apparatus, image expansion apparatus, and methods and programs thereof |
| CN105635624A (en) * | 2014-10-27 | 2016-06-01 | 华为技术有限公司 | Video image processing method, device and system |
| CN108475420A (en) * | 2015-11-16 | 2018-08-31 | 阿尔卡特朗讯美国公司 | Multiresolution compressed sensing image procossing |
| CN106791854A (en) * | 2016-11-22 | 2017-05-31 | 北京疯景科技有限公司 | Image Coding, coding/decoding method and device |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11438564B2 (en) | 2019-02-25 | 2022-09-06 | Lumicore Microelectronics Shanghai Co. Ltd. | Apparatus and method for near-eye display based on human visual characteristics |
| WO2020173414A1 (en) * | 2019-02-25 | 2020-09-03 | 昀光微电子（上海）有限公司 | Human vision characteristic-based near-eye display method and device |
| CN111698453B (en) * | 2019-03-11 | 2022-02-08 | 杭州海康威视系统技术有限公司 | Video processing method and device |
| CN111698453A (en) * | 2019-03-11 | 2020-09-22 | 杭州海康威视系统技术有限公司 | Video processing method and device |
| CN111988661A (en) * | 2019-05-24 | 2020-11-24 | 米利雅得广告公开股份有限公司 | Incorporating visual objects into video material |
| CN110572656B (en) * | 2019-09-19 | 2021-11-19 | 江苏视博云信息技术有限公司 | Encoding method, image processing method, device, system, storage medium and equipment |
| CN110572656A (en) * | 2019-09-19 | 2019-12-13 | 北京视博云科技有限公司 | coding method, image processing method, device and system |
| US11968379B2 (en) | 2019-09-27 | 2024-04-23 | Tencent Technology (Shenzhen) Company Limited | Video decoding method and apparatus, video encoding method and apparatus, storage medium, and electronic device |
| WO2021057684A1 (en) * | 2019-09-27 | 2021-04-01 | 腾讯科技（深圳）有限公司 | Video decoding method and apparatus, video encoding method and apparatus, storage medium, and electronic apparatus |
| WO2021073336A1 (en) * | 2019-10-18 | 2021-04-22 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | A system and method for creating real-time video |
| CN110796997B (en) * | 2019-11-14 | 2021-12-21 | 京东方科技集团股份有限公司 | Method and device for realizing non-uniform resolution display |
| CN110796997A (en) * | 2019-11-14 | 2020-02-14 | 京东方科技集团股份有限公司 | Method and device for realizing non-uniform resolution display |
| CN113518249A (en) * | 2020-04-10 | 2021-10-19 | 华为技术有限公司 | Far-end image processing method and device |
| CN112884803A (en) * | 2020-08-18 | 2021-06-01 | 眸芯科技（上海）有限公司 | Real-time intelligent monitoring target detection method and device based on DSP |
| CN112884803B (en) * | 2020-08-18 | 2023-11-24 | 眸芯科技（上海）有限公司 | Real-time intelligent monitoring target detection method and device based on DSP |
| CN113630622A (en) * | 2021-06-18 | 2021-11-09 | 中图云创智能科技(北京)有限公司 | Panoramic video image processing method, server, target device, apparatus and system |
| CN113630622B (en) * | 2021-06-18 | 2024-04-26 | 中图云创智能科技(北京)有限公司 | Panoramic video image processing method, server, target equipment, device and system |
| CN113592712A (en) * | 2021-07-30 | 2021-11-02 | 北京百度网讯科技有限公司 | Image processing method, device, equipment, storage medium and cloud VR system |
| CN114274877B (en) * | 2021-11-29 | 2023-06-23 | 惠州华阳通用智慧车载系统开发有限公司 | Rearview mirror image display control method |
| CN114274877A (en) * | 2021-11-29 | 2022-04-05 | 惠州华阳通用智慧车载系统开发有限公司 | Rearview mirror image display control method |
| CN114520873A (en) * | 2021-12-31 | 2022-05-20 | 广州文远知行科技有限公司 | Sweeper video acceleration method and system |
| CN116761019A (en) * | 2023-08-24 | 2023-09-15 | 瀚博半导体(上海)有限公司 | Video processing method, system, computer device and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109194923B (en) | 2023-08-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109194923A (en) | Video image processing apparatus, system and method based on non-uniformed resolution ratio | |
| US6829391B2 (en) | Adaptive resolution system and method for providing efficient low bit rate transmission of image data for distributed applications | |
| US7181046B2 (en) | Person tagging in an image processing system utilizing a statistical model based on both appearance and geometric features | |
| US8199165B2 (en) | Methods and systems for object segmentation in digital images | |
| CN109299703B (en) | Method and device for carrying out statistics on mouse conditions and image acquisition equipment | |
| US8264524B1 (en) | System for streaming multiple regions deriving from a wide-angle camera | |
| CN116018616A (en) | Maintaining a fixed size of a target object in a frame | |
| EP1416734B1 (en) | Event detection for video surveillance systems using transform coefficients of compressed images | |
| JP2004023373A (en) | Image processor and processing method, and computer program and computer-readable storage medium | |
| JP2010009134A (en) | Image processing system, image processing method, and program | |
| US9576335B2 (en) | Method, device, and computer program for reducing the resolution of an input image | |
| CN105868748A (en) | Data service platform on basis of cloud processing | |
| CN111062234A (en) | Monitoring method, intelligent terminal and computer readable storage medium | |
| CN112365407B (en) | Panoramic stitching method for camera with configurable visual angle | |
| CN113012383B (en) | Fire detection alarm method, related system, related equipment and storage medium | |
| CN108717704B (en) | Target tracking method based on fisheye image, computer device and computer readable storage medium | |
| CN112104869A (en) | Video big data storage and transcoding optimization system | |
| CN114140745A (en) | Method, system, device and medium for detecting personnel attributes of construction site | |
| CN202488602U (en) | All-dimensional vision detection analysis prewarning system based on fisheye correction technology | |
| US8600157B2 (en) | Method, system and computer program product for object color correction | |
| Zhang et al. | Nonlocal edge-directed interpolation | |
| WO2021248564A1 (en) | Panoramic big data application monitoring and control system | |
| CN112118421A (en) | Intelligent river patrol method, device, storage medium and electronic equipment | |
| CN102638674A (en) | Omnidirectional visual detecting, analyzing and warning system on basis of fisheye correction technology | |
| CA2450875A1 (en) | Remote video recognition system |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 201210 floor 8, building 1, No. 298, Xiangke Road, Pudong New Area, Shanghai Applicant after: MOUXIN TECHNOLOGY (SHANGHAI) Co.,Ltd. Address before: 201210 Room 502, building 1, No. 800, Naxian Road, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai Applicant before: MOUXIN TECHNOLOGY (SHANGHAI) Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |