CN106899810A - A kind of mine video image fusion method and device - Google Patents
A kind of mine video image fusion method and device Download PDFInfo
- Publication number
- CN106899810A CN106899810A CN201510952077.3A CN201510952077A CN106899810A CN 106899810 A CN106899810 A CN 106899810A CN 201510952077 A CN201510952077 A CN 201510952077A CN 106899810 A CN106899810 A CN 106899810A
- Authority
- CN
- China
- Prior art keywords
- image
- mine
- module
- block
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 12
- 230000004927 fusion Effects 0.000 claims abstract description 43
- 230000006835 compression Effects 0.000 claims abstract description 23
- 238000007906 compression Methods 0.000 claims abstract description 23
- 238000005070 sampling Methods 0.000 claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 11
- 239000011159 matrix material Substances 0.000 claims description 65
- 239000013598 vector Substances 0.000 claims description 22
- 238000004891 communication Methods 0.000 claims description 9
- 230000006854 communication Effects 0.000 claims description 9
- 238000011084 recovery Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 8
- 230000015572 biosynthetic process Effects 0.000 claims description 6
- 238000003786 synthesis reaction Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 230000010354 integration Effects 0.000 claims description 4
- 230000008447 perception Effects 0.000 claims description 4
- 238000000638 solvent extraction Methods 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 239000000155 melt Substances 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims 1
- 230000005540 biological transmission Effects 0.000 abstract description 8
- 230000000903 blocking effect Effects 0.000 abstract description 2
- 230000000875 corresponding effect Effects 0.000 description 10
- 238000002156 mixing Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000006837 decompression Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/222—Studio circuitry; Studio devices; Studio equipment
- H04N5/262—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects
- H04N5/265—Mixing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- 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
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a kind of mine video image fusion method, sampling, fusion and reconstruct for realizing mine image.The method to mine video image blocking compression sampling, then carries out information source entropy fusion to tile compression sampled signal first, recovers to become original image after finally the signal for merging is reconstructed, is spliced.The present invention further discloses a kind of mine video image fusing device, the device includes mine image capture device, mine image co-registration node and mine image receiving apparatus.The device uses wireless transmission method, image Compression and transmittability are strong, the redundant computation of Computer Vision is few, required memory space is small, downhole transmitted channel bandwidth requirement is low, meet the specific use environment in underground, effectively increase compression treatment and the transmittability of downhole video image, it is ensured that the real-time transmission of vision signal under mine WSN, Zigbee wireless network environment.
Description
Technical field
The present invention relates to a kind of radio communication and image fusion technology, specifically, it is related to a kind of mine video image fusion method
With device.
Background technology
Existing mine image information fusion technology needs to consider all pixels point of image, that is, needs to believe the image of compressed encoding
It is merged again after breath decompression recovery, there is complicated design, transmission bandwidth requirement height, memory space using this technique device
The problems such as demand is big.Especially, video image is influenceed larger by noise jamming in Minepit environment, is existed in mine video and is differentiated
The phenomenon such as rate is low, image blurring, but because Mine Monitoring and video communication are higher to mine video image quality and requirement of real-time,
In addition underground communica tion environment bandwidth resource-constrained, is difficult to solve to go out when video image compression is processed using traditional image fusion technology
Existing video signal transmission postpones, flating the problems such as.Therefore, in the urgent need to studying and inventing a kind of new mine video figure
As fusion method and device, to solve Problems Existing.
2006, Donoho etc. proposed compressed sensing (Compressed Sensing, CS) theory, by signal Analysis sheet
The sparse matrix of body, it is intended to break through the bandwidth bottleneck in conventional information opinion, very big carrying is generated to signal transacting and compression limit
Rise.The theory shows, if signal is compressible or is sparse in certain transform domain, then with one and can just convert
The incoherent random observation matrix of base projects on a lower dimensional space dimensional signal high obtained by conversion, then by solving one
Sparse optimization problem realizes the Accurate Reconstruction of signal, can use the sampling number reconstruction signal far below sampling thheorem requirement.And
And, there are some researches show use information entropy enters the fusion of row information as weight coefficient in perception domain, can intactly reconstruct artwork
Picture, can reduce the requirement of transmission performance, the complexity of hardware design and the demand of memory space.
Therefore, theoretical based on above-mentioned CS, research rope is based on the mine video image integration technology and method of compressive sensing theory,
By for mine wireless network environment go down into a mine wireless video image compression treatment provide a new approaches.
The content of the invention
Present invention mainly solves the problems of prior art, a kind of mine video image perceived based on splits' positions has been invented
Fusion method and device.Splits' positions are perceived the compression sampling and image for applying to mine video image with information source entropy blending algorithm
Fusion, the algorithm perceived using splits' positions, reduces the sampling rate of signal and the complexity of random observation matrix, so as to subtract
The redundant computation of few Computer Vision, saving memory space；Using information source entropy blending algorithm, can be real before Image Reconstruction
Existing information fusion, so as to reduce the requirement to transmission channel bandwidth.
The technical solution adopted by the present invention is：A kind of mine video image fusion method and device, are perceived using based on splits' positions
Information source entropy blending algorithm, for realize mine video signal sampling, fusion and reconstruct.The mine video image melts
Conjunction method, comprises the following steps：
The sampling of step 1, signal, including following sub-step：
1.1) the mine video signal of input is divided into the S image block of size K × K pixels, wherein S, K ∈ N；
1.2) build s (s=1,2 ... S) the corresponding observing matrix of individual image block is ΦK, wherein,ms＜ K2,
And ΦKIt is Hadamard matrixes；
1.3) the different observation vectors for obtaining s-th image block are respectively x1=ΦKΘ1And x2=ΦKΘ2, whereinAnd Θ1, Θ2The respectively column vector of source images s blocks；
The fusion of step 2, signal, including following sub-step：
2.1) calculating observation vector x1And x2Information source entropy entropy (x1) and entropy (x2), combination entropy entropy (x1, x2), mutually
Information content MI (x1, x2),
2.2) calculating observation vector x1And x2Weight coefficient,
2.3) it is final observation x to merge different observation vectorsb=t1x1+t2x2, wherein, t1, t2By normalized,
Meet t1+t2=1；
The reconstruct of step 3, signal, according to the corresponding perception matrix of image block signal, using orthogonal matching pursuit algorithm reconstruct image
As the corresponding reconstruction value of block signal；
The integration of step 4, signal, the method according to being divided to source images block, the corresponding reconstruction value of each block is spliced,
Reform, obtain reconstructed image.
The present invention further discloses a kind of mine video image fusing device based on compressed sensing, the mine video is applied to
Image interfusion method, the device includes：Mine image capture device, mine image co-registration node and mine image receiving apparatus；
Wherein, the mine image capture device, for the collection of mine image block, image compression encoding；The mine image co-registration
Node is used to perceive the fusion of domain epigraph；The mine image receiving apparatus be used for mine Image Reconstruction decoding, Video processing and
Image shows；The mine image capture device, mine image co-registration node and mine image receiving apparatus are connect by radio communication
Mouth connection, for being transferred to mine image receiving apparatus after the mine picture signal fusion to gathering；The wireless communication interface is adopted
With Zigbee, WiFi or/and WCDMA, WiMAX or/and LTE air interfaces.
The mine image capture device includes：Video acquisition unit and image coding unit；The video acquisition unit is used for continuous
The piecemeal collection of image, including image block module and image capture module；
Described image piecemeal module is used for mine fragmental image processing.
Described image acquisition module is used to gathering continuous block image and the block image information transfer that will collect is to Image Coding
Unit is compressed coding.
Described image coding unit, for being compressed coding, including matrix module, storage to the block image information for collecting
Module and multiplier module.
The matrix module is used to generate compression observing matrix.
The memory module is used for the compression observing matrix of memory matrix module generation.
The multiplier module is used for the compression observation square of the blocked image signal and memory module storage for collecting video acquisition unit
Battle array is multiplied, to obtain the coding of the block image after compressed encoding.
The mine image co-registration node includes information source entropy computing unit and weighting fusion treatment unit；The information source entropy computing unit,
For the calculating of correlated source entropy, including information source entropy module, combination entropy module, mutual information module and memory module；
The information source entropy module, combination entropy module and mutual information module are respectively used to calculate information source entropy, the connection of piecemeal collection signal
Close entropy and mutual information.
The memory module is used for information source entropy, combination entropy and the mutual information of memory partitioning collection signal.
The weighting fusion treatment unit, for the weighting fusion treatment to sampled signal, including power computing module and Fusion Module.
The power computing module is used to calculate the weight coefficient that piecemeal gathers image information.
The Fusion Module is used for the weighting fusion treatment that piecemeal gathers image information.
The mine image receiving apparatus include image reconstruction unit and Video Composition unit；Described image reconfiguration unit, for right
Image after splits' positions fusion carries out decoding recovery, including matrix module, memory module, multiplier module, correction module and block
Synthesis module.
The matrix module is used to generate sparse expression matrix；
The memory module is used for the sparse expression matrix of memory matrix module generation；
The multiplier module is used to calculate the orthogonal basic matrix of the Image Coding after splits' positions coding and matrix sparse expression matrix；
The correction module is used to calculate the noise residual error of the image after orthogonal basic matrix is encoded with splits' positions, and iterate to calculate,
Correction obtains sparse decoded recovery image；
Described piece of synthesis module is used for the splicing of block image information, is reformatted into original image；
The Video Composition unit is used to for the recovery image after sparse coding to merge into continuous videos.
The mine video image fusing device includes wireless camera, vehicle-mounted mobile platform and communication equipment etc. hand held mobile station.
The mine image capture device, mine image co-registration node and mine image receiving apparatus are inbeing safe explosion prevention device.
The beneficial effects of the present invention are：
The present invention is using splits' positions sampling, and required random observation matrix is simple, effectively reduces Image Coding with compression treatment
Redundant computation amount and storage demand；Especially with mine video image fusion method, before Image Reconstruction, line number is just entered
According to fusion, so as to reduce requirement of the mine video image to transmission channel.
Brief description of the drawings
Fig. 1 is a kind of mine video image fusion method schematic diagram；
Fig. 2 is a kind of mine video image fusion treatment schematic diagram perceived based on splits' positions；
Fig. 3 is a kind of mine video image fusing device schematic diagram；
Specific embodiment
Specific embodiment of the invention is described in detail below in conjunction with the accompanying drawings.
Reference picture 1, is a kind of mine video image fusion method schematic diagram, the sampling of its mine video signal and fusion process
It is as follows：First to original mine video image blocking, random observation is then carried out to piecemeal mine image, then observation is obtained
The splits' positions signal that signal is mine image carries out information source entropy fusion, finally fusion gained signal is reconstructed, is spliced,
Obtain original mine image.Splits' positions are perceived the compression sampling and figure for applying to mine video image with information source entropy blending algorithm
As fusion, the algorithm perceived using splits' positions reduces the sampling rate of signal and the complexity of random observation matrix, so that
The redundant computation of Computer Vision is reduced, memory space is saved；Using information source entropy blending algorithm, merged before Image Reconstruction
Information, it is intended to reduce transmission channel bandwidth requirement.
Reference picture 2, is a kind of mine video image fusion treatment schematic diagram perceived based on splits' positions.Mine video image
CS observations are a linear processes, and in order to ensure Accurate Reconstruction, the necessary and sufficient condition that system of linear equations has solution is observing matrix and dilute
Thin transformation matrix meets limited equidistant condition.The matrix constituted with most of fixed orthogonal basis in view of Hadamard matrixes not phase
Close, this characteristic determines to select it as the observing matrix of blocked image signal, while setting the corresponding sparse group moment of blocked image signal
Battle array is Ψ, thus need to build s (s=1,2 ... S) the corresponding observing matrix of individual image block is ΦK。
Mine video image fusion treatment is crossed and claims the whole of the main sampling including signal, the fusion of signal, the reconstruct of signal and signal
The steps such as conjunction.Implement step as follows：
The sampling of step 1, signal, including following sub-step：
1.1) the mine video signal of input is divided into the S image block of size K × K pixels, wherein S, K ∈ N；
1.2) build s (s=1,2 ... S) the corresponding observing matrix of individual image block is ΦK, wherein,
ms＜ K2, and ΦKIt is Hadamard matrixes；
1.3) the different observation vectors for obtaining s-th image block are respectively x1=ΦKΘ1And x2=ΦKΘ2, whereinAnd Θ1, Θ2The respectively column vector of source images s blocks；
The fusion of step 2, signal, including following sub-step：
2.1) calculating observation vector x1And x2Information source entropy entropy (x1) and entropy (x2), combination entropy entropy (x1, x2)、
Mutual information MI (x1, x2),
2.2) calculating observation vector x1And x2Weight coefficient,
2.3) it is final observation x to merge different observation vectorsb=t1x1+t2x2, wherein, t1, t2By normalized,
Meet t1+t2=1；
During above-mentioned formula is calculated, wherein, observation vector x1And x2Weight coefficient t1, t2Middle Section 1 respectively describes information source entropy
entropy(x1) and entropy (x2) in the middle ratio of combination entropy, include two public parts of information to be fused, it is necessary to pass through to subtract
Section 2 is gone to correct；Observation vector x1And x2Weight coefficient t1, t2Section 2 in MI mutual informations, reflect two and wait to melt
The information that conjunction information is jointly comprised, last renormalization t1, t2, that is, meet t1+t2=1；
2.3) it is final observation x to merge different observation vectorsb=t1x1+t2x2。
The reconstruct of step 3, signal, according to the corresponding perception matrix of image block signal, using orthogonal matching restructing algorithm reconstructed blocks
The corresponding reconstruction value of signal, algorithm known parameters are：Fusion measured value xb, observing matrix (Hadamard matrixes) ΦK, it is dilute
Basic matrix Ψ is dredged, degree of rarefication is k, noise residual error e, obeys N (0, σ2), concrete signal reconstruct flow is as follows：
3.1) initiation parameter：Estimate signalPerceive matrix A=ΦKΨ, the indexed set of selected column vectorIteration
Number of times t=1, noise residual error e0=xb, threshold residual value Θthreshold=| | et-et-1||2；
3.2) λ is calculatedt=argjmax|<et-1, aj>|, that is, perceive the column vector a of matrix AjCoefficient correlation is most between noise residual error e
Big column vector index λ；
3.3) indexed set Γ is updatedt=Γt-1∪{λt, while the selected column vector of record
3.4) sparse signal is asked to estimateAnd make
3.5) residual error is updatedT=t+1；
3.6) judge whether to meet iteration stopping condition.As t ＞ k, if or in the presence of | | et-et-1||2＜ ξ, when | | et||2＜ ΘthresholdWhen
Iteration terminates, otherwise return to step 3.2) continue iteration, solve optimal solution, until meeting condition untill.
The integration of step 4, signal, Fig. 2 embodiments explanation, after s image block signal reconstruction is completed, according to original
The method that image block is divided, the corresponding reconstruction value of each block is spliced, is reformed, and obtains reconstructed image.
Reference picture 3, is a kind of mine video image fusing device schematic diagram, and the device mainly includes：Mine image capture device
(10), mine image co-registration node (20) and mine image receiving apparatus (30)；Wherein, mine image capture device (10),
For image block collection, image compression encoding；Mine image co-registration node (20) is for perceiving the fusion of domain epigraph；Ore deposit
Borehole image receiving device (30) is decoded for Image Reconstruction, Video processing and image show.
A kind of mine video image fusing device shown in Fig. 3, mine image capture device (10) includes：Video acquisition unit
(101), image coding unit (102)；Wherein, video acquisition unit (101) is gathered for the piecemeal of consecutive image, including
Image block module (101A) and image capture module (101B)；Image block module (101A) is used at mine image block
Reason；Image capture module (101B) is used to gather continuous block image, and the block image information transfer that will be collected is to scheming
As coding unit (102) is compressed coding.Image coding unit (102), for being carried out to the block image information for collecting
Compressed encoding, including matrix module (102A), memory module (102B) and multiplier module (102C)；Matrix module (102A)
For generating compression observing matrix；Memory module (102B) is used for the compression observing matrix of memory matrix module generation；Multiplication modulo
Block (102C) is used for the compression observation of block image and memory module (102B) storage for collecting video acquisition unit (101)
Matrix multiple, to obtain the coding of the block image after compressed encoding.
A kind of mine video image fusing device shown in Fig. 3, mine image co-registration node (20) includes information source entropy computing unit
And weighting fusion treatment unit (202) (201)；Wherein, information source entropy computing unit (201) includes information source entropy module (201A), connection
Close entropy module (201B), mutual information module (201C) and memory module (201D)；Information source entropy module (201A), combination entropy module
(201B) and mutual information module (201C) are respectively used to calculate information source entropy, combination entropy and the mutual information of piecemeal collection signal；Deposit
Storage module (201D) is used for information source entropy, combination entropy and the mutual information of memory partitioning collection signal.Weighting fusion treatment unit (202)
Including power computing module (202A) and Fusion Module (202B)；Power computing module (202A) calculates power by the comentropy of block image
Coefficient；Block image information is calculated gained weight coefficient and is weighted fusion treatment by Fusion Module (202B) according to power computing module.
A kind of mine video image fusing device shown in Fig. 3, mine image receiving apparatus (30) include image reconstruction unit (301)
With Video Composition unit (302)；Wherein, image reconstruction unit (301), for being merged to splits' positions after image decode
Recover, including the synthesis of matrix module (301A), memory module (301B), multiplier module (301C), correction module (301D) and block
Module (301E)；Wherein, matrix module (301A) is used to generate sparse expression matrix；Memory module (301B) is used for storage matrix
The sparse expression matrix of module generation；The Image Coding that multiplier module (301C) is used to calculate after splits' positions coding is sparse with matrix
The orthogonal basic matrix of expression matrix；Correction module (301D) is used to calculate making an uproar for the image after orthogonal basic matrix is encoded with splits' positions
Sound residual error, and iterate to calculate, correct and obtain sparse decoded recovery image；Block synthesis module (301E) is used for block image
Information splices, is reformatted into original image.Video Composition unit (302) merges into continuous videos for the recovery image after sparse coding.
Splits' positions are perceived the present invention compression sampling and image co-registration for applying to mine video image with information source entropy blending algorithm,
The algorithm perceived using splits' positions, reduces the sampling rate of signal and the complexity of random observation matrix, so as to reduce video
The redundant computation of image procossing, saves memory space；Using information source entropy blending algorithm, the fuse information before Image Reconstruction, purport
Reducing transmission channel bandwidth requirement.With reference to the mine video image fusing device that the inventive method is provided, colliery is disclosure satisfy that
The specific use environment in underground and demand for security.
Above content is to combine the further description that specific preferred embodiment mode is done to the present invention, it is impossible to assert this hair
Bright specific embodiment is only limitted to this, for general technical staff of the technical field of the invention, is not departing from this hair
On the premise of bright mentality of designing, some simple replacements and change can be also carried out, should all be considered as belonging to the power that the present invention is submitted to
Protection domain involved by sharp claim.
Claims (4)
1. a kind of mine video image fusion method, sampling, fusion and reconstruct for realizing mine video signal, it is special
Levy and be, comprise the following steps：
The sampling of step 1, signal, including following sub-step：
1.1) the mine video signal of input is divided into the S image block of size K × K pixels, wherein S, K ∈ N；
1.2) build s (s=1,2 ... S) the corresponding observing matrix of individual image block is ΦK, wherein,
ms＜ K2, and ΦKIt is Hadamard matrixes；
1.3) the different observation vectors for obtaining s-th image block are respectively x1=ΦKΘ1And x2=ΦKΘ2, whereinAnd Θ1, Θ2The respectively column vector of source images s blocks；
The fusion of step 2, signal, including following sub-step：
2.1) calculating observation vector x1And x2Information source entropy entropy (x1) and entropy (x2), combination entropy entropy (x1, x2)、
Mutual information MI (x1, x2),
2.2) calculating observation vector x1And x2Weight coefficient,
2.3) it is final observation x to merge different observation vectorsb=t1x1+t2x2, wherein, t1, t2At normalization
Reason, meets t1+t2=1；
The reconstruct of step 3, signal, according to the corresponding perception matrix of image block signal, using orthogonal matching pursuit algorithm reconstruct image
As the corresponding reconstruction value of block signal；
The integration of step 4, signal, the method according to being divided to source images block, the corresponding reconstruction value of each block is spliced,
Reform, obtain reconstructed image.
2. a kind of mine video image fusing device based on compressed sensing, is applied to the mine video image fusion method, its
It is characterised by, including：Mine image capture device (10), mine image co-registration node (20) and mine image receiving apparatus (30)；
Wherein, the mine image capture device (10), for the collection of mine image block, image compression encoding；The mine image
Aggregators (20) are for perceiving the fusion of domain epigraph；The mine image receiving apparatus (30) for Image Reconstruction decode,
Video processing and image show；The mine image capture device (10), mine image co-registration node (20) and mine image connect
Receiving unit (30) is connected by wireless communication interface, is connect for being transferred to mine image after the mine picture signal fusion by collection
Receiving unit (30)；It is further characterized in that,
The wireless communication interface uses Zigbee, WiFi or/and WCDMA, WiMAX or/and LTE air interfaces；And
The mine image capture device (10) includes：Video acquisition unit (101), image coding unit (102)；It is described to regard
Frequency collecting unit (101) is gathered for the piecemeal of consecutive image, including image block module (101A) and image capture module
(101B)；Described image piecemeal module (101A) is used for mine fragmental image processing；Described image acquisition module (101B) is used
In the block image information transfer for gathering continuous block image and will collect coding is compressed to image coding unit (102)；
Described image coding unit (102), for being compressed coding to the block image information for collecting, including matrix module (102A),
Memory module (102B) and multiplier module (102C)；The matrix module (102A) is used to generate compression observing matrix；It is described
Memory module (102B) is used for the compression observing matrix of memory matrix module generation；The multiplier module (102C) is used to regard
The compression observing matrix that the block image that frequency collecting unit (101) is collected is stored with memory module (102B) is multiplied, to be pressed
Reduce the staff the coding of the block image after code；
The mine image co-registration node (20) includes information source entropy computing unit (201), weighting fusion treatment unit (202)；It is described
Information source entropy computing unit (201) is including information source entropy module (201A), combination entropy module (201B), mutual information module (201C) and deposits
Storage module (201D)；The information source entropy module (201A), combination entropy module (201B) and mutual information module (201C) are respectively used to
Calculate information source entropy, combination entropy and the mutual information of each piecemeal collection signal；The memory module (201D) gathers for memory partitioning
The information source entropy of signal, combination entropy and mutual information；The weighting fusion treatment unit (202) is including power computing module (202A) and melts
Matched moulds block (202B)；Power computing module (202A) calculates weight coefficient by the comentropy of block image；The Fusion Module
Block image information is calculated gained weight coefficient and is weighted fusion treatment by (202B) according to power computing module；
The mine image receiving apparatus (30) include image reconstruction unit (301) and Video Composition unit (302)；Described image weight
Structure unit (301), for being merged to splits' positions after image carry out decoding recovery, including matrix module (301A), memory module
(301B), multiplier module (301C), correction module (301D) and block synthesis module (301E)；Wherein, the matrix module (301A)
For generating sparse expression matrix；The memory module (301B) is used for the sparse expression matrix of memory matrix module generation；It is described
Multiplier module (301C) is used to calculate the orthogonal basic matrix of the Image Coding after splits' positions coding and matrix sparse expression matrix；Institute
Correction module (301D) is stated for calculating the noise residual error of the image after orthogonal basic matrix is encoded with splits' positions, and iterate to calculate,
Correction obtains sparse decoded recovery image；Described piece of synthesis module (301E) is used for the splicing of block image information, reforms
Into original image；The Video Composition unit (302) by the recovery image after sparse coding for merging into continuous videos.
3. mine video image fusing device according to claim 2 includes wireless camera, vehicle-mounted mobile platform and hand-held shifting
The communication equipments such as dynamic platform.
4. mine image capture device according to claim 2, mine image co-registration node and mine image receiving apparatus,
Characterized in that, the mine image capture device, mine image co-registration node and mine image receiving apparatus are essential safe type
Explosion-protection equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510952077.3A CN106899810A (en) | 2015-12-21 | 2015-12-21 | A kind of mine video image fusion method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510952077.3A CN106899810A (en) | 2015-12-21 | 2015-12-21 | A kind of mine video image fusion method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106899810A true CN106899810A (en) | 2017-06-27 |
Family
ID=59189626
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510952077.3A Pending CN106899810A (en) | 2015-12-21 | 2015-12-21 | A kind of mine video image fusion method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106899810A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108782312A (en) * | 2018-03-22 | 2018-11-13 | 杨明 | Sewage disposal system based on image procossing |
| CN109102461A (en) * | 2018-06-15 | 2018-12-28 | 深圳大学 | Image reconstructing method, device, equipment and the medium of low sampling splits' positions perception |
| CN109167973A (en) * | 2018-10-24 | 2019-01-08 | 国网江苏省电力有限公司盐城供电分公司 | A kind of image transfer method and device of intelligent safety helmet |
| CN109447946A (en) * | 2018-09-26 | 2019-03-08 | 中睿通信规划设计有限公司 | A kind of Overhead optical cable method for detecting abnormality |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101236598A (en) * | 2007-12-28 | 2008-08-06 | 北京交通大学 | Independent component analysis human face recognition method based on multi- scale total variation based quotient image |
| CN101887581A (en) * | 2010-06-17 | 2010-11-17 | 东软集团股份有限公司 | Image fusion method and device |
| CN103559696A (en) * | 2013-09-29 | 2014-02-05 | 南京邮电大学 | Image fusion method based on compressed sensing |
| WO2014144246A1 (en) * | 2013-03-15 | 2014-09-18 | Cyberricade, Inc. | Cyber security |
| CN104811664A (en) * | 2015-04-27 | 2015-07-29 | 中国矿业大学(北京) | Mine video image compression and transmission method and apparatus |
-
2015
- 2015-12-21 CN CN201510952077.3A patent/CN106899810A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101236598A (en) * | 2007-12-28 | 2008-08-06 | 北京交通大学 | Independent component analysis human face recognition method based on multi- scale total variation based quotient image |
| CN101887581A (en) * | 2010-06-17 | 2010-11-17 | 东软集团股份有限公司 | Image fusion method and device |
| WO2014144246A1 (en) * | 2013-03-15 | 2014-09-18 | Cyberricade, Inc. | Cyber security |
| CN103559696A (en) * | 2013-09-29 | 2014-02-05 | 南京邮电大学 | Image fusion method based on compressed sensing |
| CN104811664A (en) * | 2015-04-27 | 2015-07-29 | 中国矿业大学(北京) | Mine video image compression and transmission method and apparatus |
Non-Patent Citations (1)
| Title |
|---|
| LUO，XIAOYAN等: "IMAGE FUSION IN COMPRESSED SENSING" * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108782312A (en) * | 2018-03-22 | 2018-11-13 | 杨明 | Sewage disposal system based on image procossing |
| CN108782312B (en) * | 2018-03-22 | 2020-10-16 | 诸暨易阳环保科技有限公司 | Sewage treatment system based on image processing |
| CN109102461A (en) * | 2018-06-15 | 2018-12-28 | 深圳大学 | Image reconstructing method, device, equipment and the medium of low sampling splits' positions perception |
| CN109447946A (en) * | 2018-09-26 | 2019-03-08 | 中睿通信规划设计有限公司 | A kind of Overhead optical cable method for detecting abnormality |
| CN109447946B (en) * | 2018-09-26 | 2021-09-07 | 中睿通信规划设计有限公司 | Overhead communication optical cable abnormality detection method |
| CN109167973A (en) * | 2018-10-24 | 2019-01-08 | 国网江苏省电力有限公司盐城供电分公司 | A kind of image transfer method and device of intelligent safety helmet |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021093393A1 (en) | Video compressed sensing and reconstruction method and apparatus based on deep neural network | |
| CN106899810A (en) | A kind of mine video image fusion method and device | |
| CN110099280B (en) | Video service quality enhancement method under limitation of wireless self-organizing network bandwidth | |
| CN104284190B (en) | Compressed image steganography encoding method based on AMBTC high-low mean value optimization | |
| CN102630011A (en) | Compressive perceptual coding and decoding method and system in video sensor network | |
| CN102176779B (en) | Wireless multimedia sensing network video signal adaptive sampling and spectrum allocation method | |
| CN104952039B (en) | Distributed image compressed sensing method for reconstructing | |
| TW201021459A (en) | Method of data transmission | |
| CN113822954B (en) | Deep learning image coding method for man-machine cooperative scene under resource constraint | |
| CN103974268B (en) | The adjustable low delay sensor network data transmission method of fine granulation | |
| WO2022217506A1 (en) | Channel information feedback method, sending end device, and receiving end device | |
| CN105825530B (en) | Littoral zone high spectrum image distribution lossy coding and coding/decoding method based on area-of-interest | |
| US10904579B2 (en) | Method and apparatus for annealing iterative geometry smoothing | |
| CN107124612A (en) | The method for compressing high spectrum image perceived based on distributed compression | |
| CN105354867A (en) | Hyperspectral image compression algorithm research of adaptive redundant dictionary compressed sensing | |
| CN102821303A (en) | Network real-time graded compression and transmission method for pixel-level video information | |
| CN102523452A (en) | Method for conversion, compression and transmission of images | |
| CN107027028A (en) | Random offset based on JND quantifies the method and system of multiple description coded decoding | |
| CN106961607A (en) | Time-domain lapped transform based on JND is multiple description coded, decoding method and system | |
| CN104683818A (en) | Image compression method based on biorthogonal invariant set multi-wavelets | |
| CN104683814B (en) | Image transfer method towards visual perception's mass and device | |
| Ragab et al. | Low Complexity Encoder with Multilabel Classification and Image Captioning Model. | |
| CN110390636A (en) | Unmanned plane is super, the simulation Zooming method of high definition picture or video data | |
| Ye et al. | Lightweight Generative Joint Source-Channel Coding for Semantic Image Transmission with Compressed Conditional GANs | |
| CN104156179A (en) | Device and method for visible light communication |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170627 |
|
| WD01 | Invention patent application deemed withdrawn after publication |