CN106470323A - The storage method of video data and equipment - Google Patents
The storage method of video data and equipment Download PDFInfo
- Publication number
- CN106470323A CN106470323A CN201510501686.7A CN201510501686A CN106470323A CN 106470323 A CN106470323 A CN 106470323A CN 201510501686 A CN201510501686 A CN 201510501686A CN 106470323 A CN106470323 A CN 106470323A
- Authority
- CN
- China
- Prior art keywords
- frame
- frames
- conterminous
- data segment
- similarity
- 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
Landscapes
- Television Signal Processing For Recording (AREA)
Abstract
The invention discloses a kind of storage method of video data and equipment, realize storing as far as possible big video data with the memory space trying one's best little.The method includes：The continuous data segment of video is stored in the first memory module of storage device；Non-conterminous two I frames are determined from described continuous data segment according to predetermined interval；Similarity according to described non-conterminous two I frames determines takes out frame step-length；Extract a part of I frame by the use of described frame step-length of taking out as remaining I frame from described continuous data segment；Described residue I frame is stored in the second memory module of described storage device.The present invention adaptively can extract suitable I frame to reduce the storage demand of video data as far as possible from video data, extends the storage cycle of the storage device of provided capacity.
Description
Technical field
The present invention relates to the memory technology of video data, more particularly, to a kind of storage method of video data and equipment.
Background technology
The video data that video encoder is exported, is typically circulated storage by jumbo storage device.Video coding
The video data of device output is all generally continuous, no matter and the memory space of storage device is big or little, always capacity is true
Fixed.Therefore, for the storage device determining, comparatively the video data that can store has one relatively
The upper limit determining.
With the development of photographic head hardware and neighboring technology, video definition etc. is also and then lifted, the video of same time period
Data is increasing, and required memory space is also increasing.Although with the continuous progress of hardware technology, storage device
Capacity also quickly riseing, but the fast development of social life, the storage demand of video data is more and more extensive, storage
The Development Technology of equipment has been difficult to keep up with the fast development of video data storage demand and fine definition etc..Generally by
Increase more storage devices to tackle the storage demand of the video data of rapid growth.
For one piece of storage device, generally when storage device residual memory space is less, it is to be covered by default circulation
Lid strategy deletes the video data of relatively early storage, the video data that after vacateing memory space, storage receives recently.As how tried one's best
The as far as possible big video data of little memory space storage, extends the storage cycle of video data in other words as far as possible, is video data
One of important technological problems urgently to be resolved hurrily in technical field of memory.
Content of the invention
The technical problem to be solved is to propose a kind of new video storage technology, realizing with little storage of trying one's best
The as far as possible big video data of space storage.
The invention provides a kind of storage method of video data, including：The continuous data segment of video is stored storage device
The first memory module in；Non-conterminous two I frames are determined from described continuous data segment according to predetermined interval；According to described
The similarity of non-conterminous two I frames determines takes out frame step-length；Extracted from described continuous data segment using described frame step-length of taking out
A part of I frame is as remaining I frame；Described residue I frame is stored in the second memory module of described storage device.
Preferably, frame step-length is taken out according to the similarity determination of described non-conterminous two I frames, including：Obtain described non-conterminous
Two I frames gray-scale maps；Judge described non-conterminous two I frames according to the gray-scale maps of described non-conterminous two I frames
Similarity；Take out frame step-length according to described similarity determines.
Preferably, the gray-scale maps of described non-conterminous two I frames are obtained, including：Reduce described non-conterminous two I frames
Resolution simultaneously retains half-tone information, obtains the gray scale thumbnail of described non-conterminous two I frames；According to described non-conterminous two
The gray-scale maps of individual I frame judge the similarity of described non-conterminous two I frames, including：According to described non-conterminous two I frames
Gray scale thumbnail judge the similarity of described non-conterminous two I frames.
Preferably, take out frame step-length according to described similarity determines, including：Described similarity is more than or equal to default similar
During degree threshold value, described predetermined interval is taken out frame step-length as described；When described similarity is less than described similarity threshold, reduce
Described predetermined interval, and from described continuous data segment, again determine described non-conterminous two according to the predetermined interval after reducing
I frame.
Preferably, non-conterminous two I frames are determined from described continuous data segment, including：The order that the reception server sends,
Described non-conterminous two I frames are determined from described continuous data segment according to described order；Or, store mould described first
When the ratio of the capacity of block free space and described memorizer is less than or equal to default proportion threshold value, from described continuous data segment
Determine described non-conterminous two I frames.
Preferably, the method includes：After extracting described a part of I frame from described continuous data segment, deposit from described first
Described continuous data segment is deleted in storage module.
Preferably, the method includes：Receive circulation cover time point, the deletion end time exists in described second memory module
Remaining I frame before described circulation cover time point.
Present invention also offers a kind of storage device of video data, including memorizer, described memorizer includes the first storage mould
Block and the second memory module, this equipment also includes：Input block, the continuous data segment of video is stored described first storage
In module, and remaining I frame is stored in described second memory module；Processor, according to predetermined interval from described continuous
Non-conterminous two I frames are determined, the similarity according to described non-conterminous two I frames determines takes out frame step-length in data segment, and
Extract a part of I frame by the use of described frame step-length of taking out as described residue I frame from described continuous data segment.
Preferably, described processor obtains the gray-scale maps of described non-conterminous two I frames, according to described non-conterminous two I
The gray-scale maps of frame judge the similarity of described non-conterminous two I frames, take out frame step-length according to described similarity determines.
Preferably, described processor reduces the resolution of described non-conterminous two I frames and retains half-tone information, obtains described
The gray scale thumbnail of non-conterminous two I frames, not phase according to the gray scale thumbnail of described non-conterminous two I frames judges
The similarity of two adjacent I frames.
Preferably, described processor, when described similarity is more than or equal to default similarity threshold, described predetermined interval is made
Take out frame step-length for described；When described similarity is less than described similarity threshold, reduce described predetermined interval, and according to reducing
Predetermined interval afterwards determines described non-conterminous two I frames again from described continuous data segment.
Preferably, described processor determines described non-conterminous two I frames according to order from described continuous data segment；Or
When the ratio of described first memory module free space and the capacity of described memorizer is less than or equal to default proportion threshold value, from
Described non-conterminous two I frames are determined in described continuous data segment；Wherein, described order by described equipment communication unit from
Server receives.
Preferably, after described processor extracts described a part of I frame from described continuous data segment, from the described first storage
Described continuous data segment is deleted in module.
Preferably, this equipment includes：Communication unit, receives circulation cover time point；Described processor is from the described second storage
Remaining I frame before described circulation cover time point for the end time is deleted in module.
Compared with prior art, the present invention reduces the amount of storage of video data by de-redundancy technology, for provided capacity
For storage device, extend the storage cycle of video data.It is suitable that the present invention can adaptively extract from video data
The storage demand to reduce video data as far as possible for the I frame.The present invention is taken out to continuous video data by de-redundancy technology
Frame unloading, reduces the storage demand of data, can efficiently extend the storage cycle of video data.The present invention is in self adaptation
Take out after frame is processed and only remain with and use I frame, not to extract I frame in regular intervals and directly to store, to embody intelligence
Can property and adaptivity.In the present invention, taking out frame process only needs for memory areas to be divided into two memory spaces, such as arranges two
Memorizer or same memorizer is divided into two memory areas, is reached by the Coutinuous store cycle and takes out frame and turn to trigger
Deposit, operation facility.In the present invention, continuous data segment is deleted immediately available as redundant data after the completion of taking out frame, and redundant data is given birth to
Deposit that the phase is short, take out frame process and make the memory space utilization rate of storage device higher.
Other features and advantages of the present invention will illustrate in the following description, and, partly become aobvious from description
And be clear to, or understood by implementing technical scheme.The purpose of the present invention and other advantages can be by explanations
In book, claims and accompanying drawing, specifically noted structure and/or flow process are realizing and to obtain.
Brief description
Accompanying drawing is used for providing further understanding of the technical scheme to the application or prior art, and constitutes one of description
Point.Wherein, the accompanying drawing of expression the embodiment of the present application is used for explaining the technical scheme of the application together with embodiments herein,
But do not constitute the restriction to technical scheme.
Fig. 1 is the schematic flow sheet of the storage method of video data of the embodiment of the present invention.
Fig. 2 is for taking out the schematic flow sheet of frame in the storage method of video data of the embodiment of the present invention.
Fig. 3 is the organigram of the storage device of video data of the embodiment of the present invention.
Specific embodiment
To describe embodiments of the present invention below with reference to drawings and Examples in detail, whereby to the present invention how application technology
Means are solving technical problem, and reach realizing process and fully understanding and implement according to this of relevant art effect.The application is real
Apply each feature in example and embodiment, can be combined with each other under the premise of not colliding, the technical scheme being formed all exists
Within protection scope of the present invention.
In addition, the step that the flow process of accompanying drawing illustrates can be held in the computer system of such as one group of computer executable instructions
OK.And although showing logical order in flow charts, but in some cases, can be with suitable different from herein
The shown or described step of sequence execution.
As shown in figure 1, the storage method of the video data of the embodiment of the present invention, two memory modules are included with storage device
As a example illustrate, the method mainly comprises the steps.
Step S110, the video data that video encoder is exported is written in the first memory module of storage device by section,
Each section is referred to as continuous data segment, and generates, for continuous data segment, the data directory area that index information is stored in storage device,
Also the index information of continuous data segment is sent to management server.Management server stores the index letter of this continuous data segment
Breath.
In embodiments of the invention, storage device includes two storage parts, is referred to as the first memory module and second
Memory module.In various embodiments, this two storage parts both can be independently arranged or same physics is deposited
The different memory spaces of memory device.First memory module is mainly used in storing the video data of video encoder output, and second deposits
Storage module is mainly used in storing the I frame data extracting from the video data of the first memory module storage.
Storage device, after network receives the video data of video encoder output, is first buffered in and divides in advance for each encoder
In the data buffer zone joined.Then pass through to parse code stream, parse each frame of video data, including I frame and non-I frame,
Again video data is pressed frame and write the first memory module.Often write a time period (such as 2 minutes or 5 minutes etc.)
Video data, storage device just generates a new continuous data segment for this.After completing the write of a continuous data segment, then
Switching continues to write next continuous data segment.
Write the time period of the first memory module, can be chosen as needed.The length of this time period determines subsequently
Take out during frame the length of redundant data vital stage and take out frame efficiency.The length of this time period is longer, and corresponding continuous data segment is just
Bigger, in section, I frame number is also more, takes out the frame time spent also more.Carry out by continuous data segment due to smoking frame, continuously
Data segment is taken out after frame completes and just can be deleted, and the time that such redundant data exists is also longer, the storage of whole storage device
Efficiency will be under some influence.Time period is longer, and the number of the continuous data segment that same data volume is generated is less, this
The continuous data segment that sample is concurrently taken out frame is also fewer, and overall to take out frame efficiency just relatively low.
When writing continuous data segment in the first memory module, generate data segment index information for each continuous data segment,
The data directory area of storage device record the segment number of each continuous data segment, storage location, the time started, the end time,
The size of each I frame and storage location in continuous data segment.The section that this data segment index information includes each continuous data segment is compiled
Number, storage location, the time started, the end time, the size of each I frame and storage location in continuous data segment.I frame
Storage location is mainly used to subsequently take out when frame is processed can quickly orient I frame, improve and take out frame efficiency.
Step S120, the index information that management server reports according to the storage device that oneself preserves, periodic detection is continuously deposited
Whether the storage cycle reaches.Each continuous data segment in first memory module has time started and the end time of oneself, leads to
Spend the end time of comparative management server current time and some continuous data segment it is possible to learn that this continuous data segment is
No reach the storage cycle.If the current time in management server deducts the end time of this continuous data segment, gained is poor
Value is more than the default Coutinuous store cycle, is considered as this continuous data segment and has reached the Coutinuous store cycle.In Coutinuous store week
When phase reaches, send to storage device and take out frame unloading order.
Step S130, storage device receives when taking out frame unloading order, takes out frame unloading order according to this, continuous data segment is entered
Row is taken out frame and is processed, and extracts I frame and carries out the de-redundancy process of I frame, then de-redundancy is processed it from continuous data segment
The remaining I frame obtaining afterwards stores in the second memory module.Storage device also deletes corresponding consecutive numbers in the first memory module
According to section, send to management server and delete message, notify management server to delete the index information of this continuous data segment.
Management server, according to this deletion message, deletes the index information of this continuous data segment.
In other examples, storage device according to the ratio of the free space of the first memory module and the capacity of itself with pre-
If proportion threshold value process to determine when to carry out to take out frame.Such as, storage device the first memory module free space with from
The ratio of the capacity of body be less than or equal to default proportion threshold value (such as 10%) when it is believed that the first memory module is remaining can
With space less it is difficult to storage larger data amount video data, now start execution determine not from continuous data segment
Two adjacent I frames simultaneously carry out follow-up frame of taking out and process.
In embodiments of the invention, when storage device carries out taking out frame process to continuous data segment, according to default de-redundancy strategy,
Extract I frame from continuous data segment, in other words remove redundancy I frame, by remove redundancy I frame after remaining I frame namely
The I frame extracting is written to the second memory module, reduces the demand to memory space for the video data.
In embodiments of the invention, when remaining I frame is write the second memory module by storage device, it is that remaining I frame is formed one
Individual I frame data section, then writes this I frame data section in the second memory module.Storage device is also this I frame data Duan Sheng
Become index information, the index information of this I frame data section is sent to management server.Management server receives this I frame index
Store after information.
Order communication unit in storage device receives the various orders that management server issues.Send out when receiving management server
Send when taking out frame unloading order, parse this and take out and take out frame time point parameter in frame unloading order.Data rope in storage device
Draw and search the end time in area and take out continuous data segment before frame time point at this, the then segment number according to continuous data segment,
The continuous data segment meeting condition is cached in frame section queue to be taken out.When taking out frame unloading, according to segment number etc. from frame to be taken out
Take out corresponding continuous data segment in section queue, this continuous data segment is carried out take out frame process.
Embodiments of the invention, the frame of taking out that can concurrently carry out multiple continuous data segments is processed.
Above-described embodiment is to include, as a example two separate memory modules, the video of the present invention to be described by storage device
The storage method of data, in other examples, this two memory modules can also coexist in same memorizer,
The different storage zone of this memorizer is considered different memory modules.That is, in technical scheme, as long as
It is memory technology continuous data segment and remaining I frame being stored separately, be all feasible.
As shown in Fig. 2 the frame of taking out that embodiments of the invention are directed to each continuous data segment is processed, mainly comprise the steps.
Step S210, according to the segment number of continuous data segment, inquires the index information of I frame in section in data directory area,
This index information includes the size of I frame and storage location in continuous data segment；Continuous data segment is obtained according to this index information
Interior all I frames.
Step S220, according to predetermined interval, determines two I frames, this two I frames exist from all I frames of continuous data segment
In all I frames of continuous data segment and non-conterminous.Such as, continuous data segment is decoded and obtain I frame sequence, with n for interval,
Two I frames just can be obtained from this I frame sequence.Subsequently just can be according to similar between this this two I frame non-conterminous
Property, to remove the redundancy I frame in continuous data segment, final acquisition can represent the surplus of whole continuous data segment for store
Remaining I frame.
Step S230, determines the whole I frame sequence from continuous data segment according to the similarity between this two I frames non-conterminous
Middle extraction I frame take out frame step-length, take out frame step-length using this and remove redundancy I frame in I frame sequence, obtain continuous data segment
Remaining I frame in whole I frame sequence.
Such as, with n for interval, the I frame of serial number k*n and (k+1) * n in decoding I frame sequence, by analyze this two
Similarity between individual I frame, it is possible to remove redundancy I frame from the whole I frame sequence of continuous data segment, stays or takes out
Take out remaining I frame.Wherein, n is predetermined interval, and k is the sequence number of I frame.
For set predetermined interval n, if the similarity of k*n and this pair of I frame of (k+1) * n is more than or equal in advance
If similarity threshold, illustrate that n less, can be taken out by this content change represented by two I frames as taking out frame step-length
Frame.If the similarity of this pair of I frame is less than this similarity threshold, this content change represented by two I frames is described relatively
Greatly, in this case using n as taking out if frame step-length carries out taking out frame, be easy for missing the information more than comparison.Now, reduce
Predetermined interval n, and aforementioned non-conterminous two I frames are again determined from continuous data segment according to predetermined interval n after reducing,
Judge again between the similarity compared between two nearer before I frames redefining, and two I frames according to new determination
Similarity come to judge whether can be using predetermined interval n after reducing as taking out frame step-length.
Specifically, obtain the gray-scale maps of this two I frames taking k*n and this pair of I frame of (k+1) * n as a example.Then basis
The gray-scale maps of this two I frames come to judge this two I frames similarity how, subsequently the similarity according to this two I frames is Lai really
Calmly extract remaining I frame from whole I frame sequence takes out frame step-length.Gray-scale maps according to this two I frames are judging this two
The similarity of I frame, mainly passes through the resolution of this two I frames of reduction and only retains the gray scale letter of the pixel of this two I frames
Breath, obtains the gray scale thumbnail of this two I frames.Subsequently similarity judgement is carried out according to gray scale thumbnail.
Specifically, after decoding I frame sequence obtains k*n and this two I frames of (k+1) * n, reduce the resolution of this two I frames
Rate is to M*N, and only retains pixel grey scale information, obtains the gray scale thumbnail of this two I frames.This gray scale thumbnail is M*N
Matrix, pixel grey scale in the corresponding thumbnail of each element in matrix, span is 0～255.
In embodiments of the invention, n=32, k=(0,1,2 ...), M*N=352*288.
According to the similarity between the gray scale thumbnail of gained, judge the similarity between this two I frames.Such as
K*n and the similarity of this two I frames of (k+1) * n, can be carried out in two steps.The first step is by judging this two I
In the gray scale thumbnail of frame, the similarity of respective pixel is judging the similarity between the gray scale thumbnail of this two I frames；Second
Step be according to the gray scale thumbnail of this two I frames between similarity judging the similarity between this two I frames.
In judging two width gray scale thumbnails during the similarity of the pixel of same position, two width gray scale thumbnails are designated as f respectively
With f ', the absolute value of the gray scale difference of the pixel in same position is abs (x, y)=| f (x, y)-f ' (x, y) |, in advance
If threshold value T.If abs (x, y)<=T is then it is assumed that this two pixels have similarity；Otherwise then it is assumed that this two
Pixel does not have similarity.Wherein, (x, y) be position in gray scale thumbnail matrix for the pixel, span be (0,
0)～(M, N).Threshold value T in the present embodiment value be 48, in other embodiments, this threshold value T also dependent on
Needs take other numerical value.
It is possible to judge two width gray scales after the correlation result of the pixel on same position in obtaining two width gray scale thumbnails
Similarity between thumbnail.Specifically, by the judgement of pixel similarity, the phase between two thumbnails can be obtained
Like property metric matrix, this matrix size is all M*N, and its element is designated as cor (x, y).Same in two width gray scale thumbnails
Pixel on position has cor (x, y) during similarity and takes 1, otherwise takes 0.Institute's similarity measurements moment matrix there is element
Summation, is designated as D, then divided by matrix size M*N, obtains thumbnail likeness coefficient r, i.e. r=D/ (M*N).Aobvious
So, r<=1.R bigger then it represents that the similarity between two width gray scale thumbnails is bigger, otherwise then similarity is less.Wherein,
X, y, M, N span refer to above.Certainly, if pixel has cor (x, y) during similarity and takes 0, instead
Take 1, then thumbnail likeness coefficient r less then it represents that thumbnail similarity is bigger, otherwise then similarity is less.
The similarity of gray scale thumbnail may act as the similarity of two I frames.
Certainly, in other embodiments of the invention, do not reduce the resolution of I frame, directly utilize the gray scale of the pixel of I frame
Information judges come the similarity to carry out two neighboring I frame, is also feasible.Same in gray-scale maps by two neighboring I frame
Similarity between pixel on position, obtains the similarity between two width gray-scale maps.Then according to the phase between two width gray-scale maps
Like property it is possible to obtain the similarity between corresponding two I frames.
According to the similarity between this two I frames, the such as similarity according to kth * n and this two I frames of (k+1) * n,
Remove the redundancy I frame in whole I frame sequence, extract acquisition is the remaining I frame in continuous data segment, these remaining I frames
This continuous data segment just can be represented.Then it is possible to the second memory module by those remaining I frame write storage devices.
In the storage method of the video data of the embodiment of the present invention, delete from the first memory module 301 at having been carried out taking out frame
The opportunity of the continuous data segment of reason is very flexible, can execute after remaining I frame is written in the second memory module 302,
Can also be remove redundancy I frame after, remaining I frame is written in the second memory module 302 before carry out.
In embodiments of the invention, finally store the remaining I frame in the second memory module, not by predetermined interval from
That a part of I frame extracting in continuous data segment, but determine extraction from I frame sequence by flexibly taking out frame step-length
Those I frames.Extract the total amount of these the remaining I frames obtaining, significantly reduce for comparing continuous data segment, significantly
Reduce the requirement to memory space.For the storage device of provided capacity, significantly extend the storage cycle of video.
Determine useful I frame by likeness coefficient r takes out frame step-length s, and r is less, and s is less, and that is, wherein redundancy I frame is got over
Few.In one embodiment of the present of invention, r and s has following corresponding relation：
r<=0.2, s=2；
0.2<r<=0.4, s=4；
0.4<r<=0.6, s=8；
0.6<r<=0.8, s=16；
0.8<R, s=24.
By the I frame between k*n and (k+1) * n, from the beginning of kth * n, with s as step-length, according to this section of I frame index
Information, the I frame meeting condition is taken out from continuous data segment and writes the second memory module, generates one after the completion of write
Corresponding I frame data section.
In embodiments of the invention, the storage device circulation cover time point that also contact mange-ment server issues.Receiving this
After the information of circulation cover time point, delete the end time from the second memory module surplus before this circulation cover time point
Remaining I frame, realizes the circulation storage of the memory space of the second memory module.
As shown in figure 3, the storage device of the video data of the embodiment of the present invention includes memorizer 300, this memorizer 300
Including the first memory module 301 and the second memory module 302, this two memory modules coexist in memorizer 300, are to deposit
The different memory area of reservoir 300.In other examples, this two memory modules can also be separate two
Physical memory devices.That is, in technical scheme, as long as continuous data segment and remaining I frame can be entered
The memory technology that row is stored separately, is all feasible.This storage device also includes input block 310 and processor 320.
The first memory module 301 in input block 310, with memorizer 300 and the second memory module 302 are respectively connected with,
Video data is stored in the first memory module 301 by section, the remaining I frame obtaining after subsequently being taken out frame stores second
In memory module 302.
Processor 320, is connected with the first memory module 301 in input block 310 and memorizer 300, decodes consecutive numbers
Obtain I frame sequence according to section, non-conterminous two I originally are obtained from this I frame sequence of continuous data segment according to predetermined interval
After frame, determine that according to the similarity between this two I frames that extracts remaining I frame from this I frame sequence takes out frame step-length, and
Using determined by take out frame step-length, extract a part of I frame from the I frame sequence of this continuous data segment as aforesaid residue
I frame.
The storage device of the video data of the embodiment of the present invention, also includes receiving management server by taking out frame unloading order transmission
The communication unit (not shown) taking out frame time point.Management server will be taken out frame time point and be encapsulated in as parameter and be taken out frame
Unloading order is sent to storage device, and the communication unit of storage device is receiving after this takes out frame unloading order, therefrom parses this
Take out frame time point.Processor 320 takes out frame time point according to this, and this storing from the first memory module 301 takes out frame time point
Described remaining I frame is extracted in continuous data segment in the past.
In other examples, the ratio of the free space according to the first memory module 301 for the processor 320 and the capacity of itself
Value to determine when with default proportion threshold value to carry out taking out frame process.Such as, processor 320 is in the first memory module 301
The ratio of free space and the capacity of itself be less than or equal to during default proportion threshold value (such as 10%) it is believed that first deposits
Storage the remaining free space of module 301 less it is difficult to storage larger data amount video data, now start execute from
Determine non-conterminous two I frames in continuous data segment and carry out follow-up frame of taking out and process.
In the storage device of the video data of the embodiment of the present invention, processor 320 extracts remaining I from continuous data segment
After frame, delete the continuous data segment having been carried out taking out frame process from the first memory module 301.Processor 320 is deposited from first
The opportunity deleting the continuous data segment having been carried out taking out frame process in storage module 301 is very flexible, can be to extract remaining I
After frame, input block 310 carry out before remaining I frame being write in the second memory module 302 of memorizer 300, also may be used
To be to carry out again after remaining I frame is written in the second memory module 302 input block 310.Continuous data segment takes out frame
After the completion of be deleted in time, life cycle is short, can improve the memory space utilization rate of the first memory module 301.
Communication unit receives management server and passes through circulation covering delete command transmission circulation cover time this parameter of point.Management
Circulation cover time point is encapsulated in circulation covering delete command as parameter and is sent to storage device, storage device by server
Communication unit receive this circulation cover delete command after, therefrom parse circulation cover time point.Processor follows according to this
Ring cover time point deletes remaining I frame before this circulation cover time point for the end time from the second memory module 302.
It is possible to vacate memory space in the second memory module 302 after deleting these remaining I frames, continue storage so that profit is follow-up
New remaining I frame.
In embodiments of the invention, de-redundancy module 330 is according in a part of I frame being extracted between two neighboring I frame
Similarity, from this part of I frame remove redundancy I frame.
In embodiments of the invention, processor 320 obtains non-conterminous two I frames from the I frame sequence of continuous data segment
After gray-scale maps, judge the similarity of this two I frames according to the gray-scale maps of this two non-conterminous I frames, further according to this two
The similarity of I frame determines takes out frame step-length.In this process, processor 320 reduces the resolution of this two I frames and retains
Respective half-tone information, obtains the gray scale thumbnail of this two I frames, further according to the gray scale thumbnail of this two I frames, to sentence
The similarity breaking between this two I frames.The similarity of this two I frames is more than or equal to default similarity threshold, then this is described
Less, processor 320 is by the spacing between this two I frames namely aforesaid default for content change represented by two I frames
Interval carries out taking out frame process as taking out frame step-length, typically will not miss important information.If this two I frames is similar
Property be less than this similarity threshold, then illustrate that this content change represented by two I frames is larger, in this case will be aforesaid pre-
If interval is as taking out if frame step-length carries out taking out frame, it is easy for missing the information more than comparison.Now, before processor 320 reduces
The predetermined interval stated so that this two I frames apart closer to thus being avoided that taking out frame omits more information.Processor 320
Again determine two I frames from continuous data segment further according to the predetermined interval after reducing, then judge to redefine compare before
Similarity between two nearer I frames, and judge whether will to contract according to the similarity between the new two I frames determining
Predetermined interval after little is as taking out frame step-length.
In embodiments of the invention, a continuous data segment is deleted after the completion of taking out frame it is possible to from the first memory module,
To vacate memory space, improve the service efficiency of whole storage device.And delete this continuous data segment from data directory area
Index information, writes the index information of newly-generated I frame data section.Delete index information and the write of this continuous data segment
The index information of I frame data section, also will report management server.
Embodiments of the invention, when reaching in the I frame storage cycle, management server sends circulation covering to storage device and deletes
Order.After storage device receives the circulation covering delete command of management server transmission, parse this circulation and cover deletion life
Circulation cover time point parameter in order.Search the end time in the data directory area of storage device in this circulation cover time point
I frame data section before, then by the end time, the I frame data section before this circulation cover time point stores mould from second
Delete in block, the I frame data section being subsequently generated with vacating space storage.
In embodiments of the invention, the segment information that storage device reports is preserved by management server, according to continuous data segment
Time started and the end time safeguard two detection times of the first memory module：First memory module record continuous data segment
Time started (TStartC) and end time (TEndC), it is corresponding to be stored in the first memory module all
The minima of the time started of continuous data segment and the maximum of end time.Management server is opened always according to I frame data section
Time beginning and end time safeguard two detection times of the second memory module：The beginning of the second memory module record I frame data section
Time (TStartI) and end time (TEndI), it is corresponding to be all I frame data that the second memory module is stored
The minima of time started of section and the maximum of end time.
Whether the management server periodic detection cycle reaches.If management server take out frame time point TCheckC (system
Current time deducts the Coutinuous store cycle) time range corresponding to continuous data segment that stored in the first memory module it
Interior, i.e. TStartC<TCheckC<TEndC, then management server issue to storage device and take out frame unloading order, in order
Contain and take out frame time point TCheckC.If (current time in system subtracts the circulation cover time point TCheckI of management server
Go to circulate covering cycle) within time range corresponding in the I frame data section that the second memory module is stored, that is,
TStartI<TCheckI<TEndI, then management server to storage device issue circulation cover delete command, contain in order
Circulation cover time point TCheckI.
In embodiments of the invention, storage device is except in the local record index information of continuous data segment and I frame data section
Index information outer it is also possible to write continuous data is generated continuous data segment, take out the I frame data section generating during frame unloading,
The continuous data segment deleted, the index information of the I frame data section of circulation covering deletion report management server.
Embodiments of the invention are applied to the Video Storage System being mainly made up of management server and storage device.Management service
Device preserves the relative index information of data storage in storage device, can also be storage device setting memory space and storage week
Phase.And, management server can also issue to storage device to take out frame unloading order and circulate covering and delete according to the storage cycle
Except order.The video data that video encoder exports is stored in the first memory module in units of section storage device, receives
When taking out frame unloading order, carry out taking out frame unloading；When receiving circulation covering delete command, it is circulated covering and deletes.
Embodiments of the invention can extend the storage cycle of video data as much as possible.
In embodiments of the invention, storage device includes the first memory module and the second memory module, updates video data
The storage cycle includes Coutinuous store cycle and I frame storage cycle.Coutinuous store cycle and I frame storage cycle sum are circulation
Covering cycle.The video data of video encoder output, the first memory module of segmentation write storage device, institute in record segment
There is I frame information, the index informations such as segment number, start and end time are reported record management server simultaneously and get off.
When reaching in the Coutinuous store cycle, management server issues to storage device takes out frame unloading order, after storage device receives
Frame unloading is taken out in execution, will decode taking out the continuous data segment interior part I frame data before frame time point in the first memory module
Gray scale thumbnail, determines according to the similarity between thumbnail and takes out frame step-length, take out the I from continuous data segment for the frame step-length using this
Remove redundancy I frame in frame sequence, write remaining useful I frame and generate I frame data section in this equipment second memory module,
Delete corresponding continuous data segment simultaneously, and report management server new segment index information and former section of deletion index information.Circulation
When covering cycle reaches, management server issues circulation to storage device and covers delete command, and storage device executes after receiving and follows
Ring cover is deleted, and I frame data section in the second memory module before circulation cover time point is deleted.
Embodiments of the invention, the way that according to the capacity of storage device in Video Storage System, video encoder can be accessed,
Code check and expected storage cycle arrange first, second memory module and Coutinuous store cycle and I frame storage cycle.
Wherein, the default storage cycle can be adjusted as needed, to be applied to different use occasions.
Specifically, the memory space of storage device and storage cycle etc., and the storage capacity of storage device between, have as follows
Corresponding relation.
Storage space S=[storage cycle T 1+ storage cycle T 2*I frame is than row coefficient] * (code check/(8*1024*1024)
* 24*3600) * video access number.
Wherein, storage space S, is system memory devices total capacity (GB).Storage cycle T 1 be the Coutinuous store cycle (with
It is unit).Storage cycle T 2 is I frame storage cycle (in units of sky).Code check is the output code of video encoder
Rate (Kbps).Video accesses the quantity that number is video encoder.I frame proportionality coefficient is I frame data capacity and continuous data
The ratio of capacity, wherein continuous data include I frame data and non-I frame data.I frame proportionality coefficient is adjustable, concrete and video
Coding I frame period is related, when extracting whole I frame, typically can use 33%, only extracts useful I frame, this value can obtain more
Little.
It is all invariant that storage space S, code check, video access way, and for user, circulation covering cycle T is as far as possible
Good a little louder, such data can preserve more long.
According to above-mentioned analysis it is assumed that T1=30 days, if not carrying out taking out frame unloading, i.e. T2=0, then T=T1=30 days.As
Fruit takes out whole I frames unloading, if T1=25 days, T2=15 days, T=T1+T2=40 days；Only pump with I frame and unloading,
Then T2>15 days, T>40 days.Thus contrast visible, for embodiments of the invention compare traditional technical scheme, permissible
Cycle stretch-out 33% will be stored even more many.
Those skilled in the art should be understood that each ingredient of the equipment that above-mentioned the embodiment of the present application is provided, and
Each step in method, they can concentrate on single computing device, or is distributed in what multiple computing devices were formed
On network.Alternatively, they can be realized with the executable program code of computing device.It is thus possible to they are stored
To be executed by computing device in the storage device, or they be fabricated to each integrated circuit modules respectively, or by they
In multiple modules or step be fabricated to single integrated circuit module to realize.So, the present invention be not restricted to any specific
Hardware and software combines.
Although disclosed herein embodiment as above, described content only readily appreciates that technical solution of the present invention is adopted
Embodiment, is not limited to the present invention.Technical staff in any art of the present invention, without departing from this
On the premise of bright disclosed spirit and scope, any modification and change can be carried out in the form implemented and details, but
The scope of patent protection of the present invention, still must be defined by the scope of which is defined in the appended claims.
Claims (14)
1. a kind of storage method of video data, including：
The continuous data segment of video is stored in the first memory module of storage device；
Non-conterminous two I frames are determined from described continuous data segment according to predetermined interval；
Similarity according to described non-conterminous two I frames determines takes out frame step-length；
Extract a part of I frame by the use of described frame step-length of taking out as remaining I frame from described continuous data segment；
Described residue I frame is stored in the second memory module of described storage device.
2. method according to claim 1, wherein, the similarity according to described non-conterminous two I frames determines to be taken out
Frame step-length, including：
Obtain the gray-scale maps of described non-conterminous two I frames；
Judge the similarity of described non-conterminous two I frames according to the gray-scale maps of described non-conterminous two I frames；
Take out frame step-length according to described similarity determines.
3. method according to claim 2, wherein：
Obtain the gray-scale maps of described non-conterminous two I frames, including：Reduce the resolution of described non-conterminous two I frames simultaneously
Retain half-tone information, obtain the gray scale thumbnail of described non-conterminous two I frames；
Judge the similarity of described non-conterminous two I frames according to the gray-scale maps of described non-conterminous two I frames, including：Root
Judge the similarity of described non-conterminous two I frames according to the gray scale thumbnail of described non-conterminous two I frames.
4. method according to claim 2, wherein, takes out frame step-length according to described similarity determines, including：
When described similarity is more than or equal to default similarity threshold, described predetermined interval is taken out frame step-length as described；
When described similarity is less than described similarity threshold, reduce described predetermined interval, and according to the predetermined interval weight after reducing
Newly determine described non-conterminous two I frames from described continuous data segment.
5. method according to claim 1, wherein, determines non-conterminous two I frames from described continuous data segment,
Including：
The order that the reception server sends, determines described non-conterminous two I according to described order from described continuous data segment
Frame；Or
It is less than or equal to default proportion threshold value in the ratio of described first memory module free space and the capacity of described memorizer
When, determine described non-conterminous two I frames from described continuous data segment.
6. method according to claim 1, wherein, the method includes：
After extracting described a part of I frame from described continuous data segment, delete described continuous from described first memory module
Data segment.
7. method according to claim 1, wherein, the method includes：
Receive circulation cover time point, in described second memory module delete the end time described circulation cover time point it
Front remaining I frame.
8. a kind of storage device of video data, including memorizer, described memorizer includes the first memory module and second and deposits
Storage module, this equipment also includes：
Input block, the continuous data segment of video is stored in described first memory module, and remaining I frame is stored
In described second memory module；
Processor, determines non-conterminous two I frames according to predetermined interval, according to described non-conterminous from described continuous data segment
The similarity of two I frames determine and take out frame step-length, and extract one from described continuous data segment using described frame step-length of taking out
Divide I frame as described residue I frame.
9. equipment according to claim 8, wherein：
Described processor obtains the gray-scale maps of described non-conterminous two I frames, according to the gray scale of described non-conterminous two I frames
Figure judges the similarity of described non-conterminous two I frames, takes out frame step-length according to described similarity determines.
10. equipment according to claim 9, wherein：
Described processor reduces the resolution of described non-conterminous two I frames and retains half-tone information, obtains described non-conterminous
The gray scale thumbnail of two I frames, judges described non-conterminous two according to the gray scale thumbnail of described non-conterminous two I frames
The similarity of I frame.
11. equipment according to claim 9, wherein：
Described processor, when described similarity is more than or equal to default similarity threshold, described predetermined interval is taken out as described
Frame step-length；When described similarity is less than described similarity threshold, reduce described predetermined interval, and according to default after reducing
Interval determines described non-conterminous two I frames again from described continuous data segment.
12. equipment according to claim 8, wherein：
Described processor determines described non-conterminous two I frames according to order from described continuous data segment；Or described
When the ratio of the capacity of one memory module free space and described memorizer is less than or equal to default proportion threshold value, from described continuous
Described non-conterminous two I frames are determined in data segment；
Wherein, described order is received from server by the communication unit of described equipment.
13. equipment according to claim 8, wherein：
After described processor extracts described a part of I frame from described continuous data segment, delete from described first memory module
Except described continuous data segment.
14. equipment according to claim 8, wherein, this equipment includes：
Communication unit, receives circulation cover time point；
Remaining I before described circulation cover time point for the end time deleted from described second memory module by described processor
Frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510501686.7A CN106470323B (en) | 2015-08-14 | 2015-08-14 | The storage method and equipment of video data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510501686.7A CN106470323B (en) | 2015-08-14 | 2015-08-14 | The storage method and equipment of video data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106470323A true CN106470323A (en) | 2017-03-01 |
| CN106470323B CN106470323B (en) | 2019-08-16 |
Family
ID=58214896
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510501686.7A Active CN106470323B (en) | 2015-08-14 | 2015-08-14 | The storage method and equipment of video data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106470323B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109816110A (en) * | 2019-01-24 | 2019-05-28 | 杭州嘉楠耘智信息科技有限公司 | Scrypt algorithm workload proving method and device |
| CN110300134A (en) * | 2018-03-22 | 2019-10-01 | 杭州海康威视系统技术有限公司 | Memory space method of adjustment, device and the cloud storage system of cloud storage resource pool |
| CN110728699A (en) * | 2019-10-11 | 2020-01-24 | 智慧视通(杭州)科技发展有限公司 | Track post-processing method based on characteristic distance |
| CN111698555A (en) * | 2020-06-19 | 2020-09-22 | 浙江大华技术股份有限公司 | Video frame extraction processing method and device |
| CN111866457A (en) * | 2020-07-14 | 2020-10-30 | 广州市宏视电子技术有限公司 | Monitoring image processing method, electronic device, storage medium and system |
| CN112434185A (en) * | 2020-10-26 | 2021-03-02 | 国家广播电视总局广播电视规划院 | Method, system, server and storage medium for searching similar video clips |
| CN112541390A (en) * | 2020-10-30 | 2021-03-23 | 四川天翼网络服务有限公司 | Frame-extracting dynamic scheduling method and system for violation analysis of examination video |
| CN112995676A (en) * | 2021-04-25 | 2021-06-18 | 北京金和网络股份有限公司 | Audio and video compression method, device, equipment, system and storage medium |
| CN114245167A (en) * | 2021-11-08 | 2022-03-25 | 浙江大华技术股份有限公司 | Video storage method and device and computer readable storage medium |
| CN114430469A (en) * | 2022-04-01 | 2022-05-03 | 浙江大华技术股份有限公司 | Video data storage method, video data reading method, electronic device and readable storage medium |
| CN115618050A (en) * | 2022-12-06 | 2023-01-17 | 苏州浪潮智能科技有限公司 | Video data storage and analysis method, device, system, communication equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070061727A1 (en) * | 2005-09-15 | 2007-03-15 | Honeywell International Inc. | Adaptive key frame extraction from video data |
| CN101674444A (en) * | 2009-10-15 | 2010-03-17 | 杭州华三通信技术有限公司 | Storage method of video data, system and encoder thereof |
| CN102761733A (en) * | 2011-04-29 | 2012-10-31 | 富泰华工业（深圳）有限公司 | Monitoring system and video data storage method |
| CN103810711A (en) * | 2014-03-03 | 2014-05-21 | 郑州日兴电子科技有限公司 | Keyframe extracting method and system for monitoring system videos |
| CN104618679A (en) * | 2015-03-13 | 2015-05-13 | 南京知乎信息科技有限公司 | Method for extracting key information frame from monitoring video |
-
2015
- 2015-08-14 CN CN201510501686.7A patent/CN106470323B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070061727A1 (en) * | 2005-09-15 | 2007-03-15 | Honeywell International Inc. | Adaptive key frame extraction from video data |
| CN101674444A (en) * | 2009-10-15 | 2010-03-17 | 杭州华三通信技术有限公司 | Storage method of video data, system and encoder thereof |
| CN102761733A (en) * | 2011-04-29 | 2012-10-31 | 富泰华工业（深圳）有限公司 | Monitoring system and video data storage method |
| CN103810711A (en) * | 2014-03-03 | 2014-05-21 | 郑州日兴电子科技有限公司 | Keyframe extracting method and system for monitoring system videos |
| CN104618679A (en) * | 2015-03-13 | 2015-05-13 | 南京知乎信息科技有限公司 | Method for extracting key information frame from monitoring video |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110300134B (en) * | 2018-03-22 | 2022-10-04 | 杭州海康威视系统技术有限公司 | Storage space adjusting method and device of cloud storage resource pool and cloud storage system |
| CN110300134A (en) * | 2018-03-22 | 2019-10-01 | 杭州海康威视系统技术有限公司 | Memory space method of adjustment, device and the cloud storage system of cloud storage resource pool |
| CN109816110A (en) * | 2019-01-24 | 2019-05-28 | 杭州嘉楠耘智信息科技有限公司 | Scrypt algorithm workload proving method and device |
| CN110728699A (en) * | 2019-10-11 | 2020-01-24 | 智慧视通(杭州)科技发展有限公司 | Track post-processing method based on characteristic distance |
| CN111698555A (en) * | 2020-06-19 | 2020-09-22 | 浙江大华技术股份有限公司 | Video frame extraction processing method and device |
| CN111866457A (en) * | 2020-07-14 | 2020-10-30 | 广州市宏视电子技术有限公司 | Monitoring image processing method, electronic device, storage medium and system |
| CN112434185A (en) * | 2020-10-26 | 2021-03-02 | 国家广播电视总局广播电视规划院 | Method, system, server and storage medium for searching similar video clips |
| CN112434185B (en) * | 2020-10-26 | 2023-07-14 | 国家广播电视总局广播电视规划院 | Method, system, server and storage medium for searching similar video clips |
| CN112541390A (en) * | 2020-10-30 | 2021-03-23 | 四川天翼网络服务有限公司 | Frame-extracting dynamic scheduling method and system for violation analysis of examination video |
| CN112541390B (en) * | 2020-10-30 | 2023-04-25 | 四川天翼网络股份有限公司 | Frame extraction dynamic scheduling method and system for examination video violation analysis |
| CN112995676A (en) * | 2021-04-25 | 2021-06-18 | 北京金和网络股份有限公司 | Audio and video compression method, device, equipment, system and storage medium |
| CN114245167A (en) * | 2021-11-08 | 2022-03-25 | 浙江大华技术股份有限公司 | Video storage method and device and computer readable storage medium |
| CN114430469A (en) * | 2022-04-01 | 2022-05-03 | 浙江大华技术股份有限公司 | Video data storage method, video data reading method, electronic device and readable storage medium |
| CN115618050A (en) * | 2022-12-06 | 2023-01-17 | 苏州浪潮智能科技有限公司 | Video data storage and analysis method, device, system, communication equipment and storage medium |
| CN115618050B (en) * | 2022-12-06 | 2023-03-21 | 苏州浪潮智能科技有限公司 | Video data storage and analysis method, device, system, communication equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106470323B (en) | 2019-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106470323A (en) | The storage method of video data and equipment | |
| CN110309837B (en) | Data processing method and image processing method based on convolutional neural network characteristic diagram | |
| US11222211B2 (en) | Method and apparatus for segmenting video object, electronic device, and storage medium | |
| Chen et al. | MICU: Image super-resolution via multi-level information compensation and U-net | |
| CN109522967A (en) | A kind of commodity attribute recognition methods, device, equipment and storage medium | |
| CN110324706B (en) | Video cover generation method and device and computer storage medium | |
| US20100322301A1 (en) | Image processor, image generator and computer program | |
| CN105550246B (en) | The system and method for load networks picture under Android platform | |
| CN110378230A (en) | Missing face identification method, device, computer equipment and storage medium | |
| CN110147469A (en) | A kind of data processing method, equipment and storage medium | |
| CN113221983B (en) | Training method and device for transfer learning model, image processing method and device | |
| CN113792742A (en) | Semantic segmentation method of remote sensing image and training method of semantic segmentation model | |
| CN111222454A (en) | Method and system for training multi-task target detection model and multi-task target detection | |
| CN104662900B (en) | video image encoding apparatus | |
| CN112308802A (en) | Image analysis method and system based on big data | |
| KR100210581B1 (en) | Image data processing method and system for giving identification labels to areas of connected black picture elements | |
| CN112188215B (en) | Video decoding method, device, equipment and storage medium | |
| CN104268169B (en) | A kind of remote sensing image data immediate processing method based on PS softwares | |
| CN104657960A (en) | Gray level image contrast drawing method and device | |
| CN104516899B (en) | Character library update method and device | |
| US11663184B2 (en) | Information processing method of grouping data, information processing system for grouping data, and non-transitory computer readable storage medium | |
| CN109274966A (en) | A kind of monitor video content De-weight method and system based on motion vector | |
| CN115861100A (en) | Video processing method, video processing device, computer equipment and storage medium | |
| CN113591838B (en) | Target detection method, device, electronic equipment and storage medium | |
| CN104462422A (en) | Object processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |