CN107103595A - Method, device, storage medium and the equipment of detection image change - Google Patents
Method, device, storage medium and the equipment of detection image change Download PDFInfo
- Publication number
- CN107103595A CN107103595A CN201710393356.XA CN201710393356A CN107103595A CN 107103595 A CN107103595 A CN 107103595A CN 201710393356 A CN201710393356 A CN 201710393356A CN 107103595 A CN107103595 A CN 107103595A
- Authority
- CN
- China
- Prior art keywords
- band
- frequency sub
- processing
- disparity map
- low frequency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 70
- 230000008859 change Effects 0.000 title claims abstract description 69
- 238000001514 detection method Methods 0.000 title claims abstract description 42
- 238000003860 storage Methods 0.000 title claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 164
- 230000002708 enhancing effect Effects 0.000 claims abstract description 23
- 230000009466 transformation Effects 0.000 claims description 40
- 238000006243 chemical reaction Methods 0.000 claims description 35
- 238000005070 sampling Methods 0.000 claims description 20
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 230000003044 adaptive effect Effects 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 8
- 238000004148 unit process Methods 0.000 claims description 2
- 238000004590 computer program Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000009467 reduction Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/37—Determination of transform parameters for the alignment of images, i.e. image registration using transform domain methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses method, device, storage medium and the equipment of a kind of detection image change, it is related to technical field of image processing, it is possible to increase the degree of accuracy of detection SAR image situation of change.The method of the present invention mainly includes：Obtain the disparity map of two width image to be detected；Frequency band parsing is carried out to the disparity map, low frequency sub-band and high-frequency sub-band is obtained；Linear enhancing processing is carried out to the low frequency sub-band, denoising is carried out to the high-frequency sub-band；Low frequency sub-band after processing and the high-frequency sub-band after processing are merged, the disparity map after being handled；Clustering processing is carried out to the disparity map after the processing, change information and non-changing information between described two width image to be detected is obtained.In the scene detected the present invention is mainly suitable for the situation of change to SAR image.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of method of detection image change, device, storage
Medium and equipment.
Background technology
With the development of science and technology, the technology of collection image is more and more, it is, for example, possible to use camera gathers image,
Radar can be used to gather image etc..In order to obtain the higher image of resolution ratio under the extremely low meteorological condition of visibility, at present
A kind of synthetic aperture radar (Synthetic Aperture Radar, SAR) is invented.Can be with round-the-clock, round-the-clock using SAR
Ground is observed to destination object (such as earth's surface), collection not in the same time under high-definition picture, then by these figures
As being analyzed, image change result is obtained.However, during SAR image is imaged, inevitably being multiplied property is made an uproar
The influence of sound, so that every width SAR image has larger difference with actual destination object, and then causes detection image change
The degree of accuracy substantially reduce.
At present, main two kinds of the method for detection image change：(1) priori (some on destination object are first obtained
Less image is influenceed by multiplicative noise), then these prioris are trained, training pattern is obtained, is finally based on instruction
Practice model to needing the image detection situation of change of detection；(2) disparity map of two images to be detected is first obtained, then to this
The pixel of disparity map is clustered, and therefrom finds out change class and non-changing class.
For method (1), because method (1) is to be become based on relatively accurate priori to analyze SAR image
Change situation, even if so SAR image to be detected is influenceed larger by multiplicative noise, also not interfering with and SAR image being changed
The result of detection, but in most cases, priori can not be got, therefore the practicality of this method is relatively low.It is right
For method (2), because method (2) is directly to be clustered by the disparity map to two images to be detected, image is obtained
Result of variations, so when this two images is influenceed larger by multiplicative noise, the accurate of Image Change Detection result can be caused
Degree is substantially reduced.Therefore, the degree of accuracy for how improving SAR image change testing result is urgently to be resolved hurrily.
The content of the invention
In view of this, method, device, storage medium and the equipment for a kind of detection image change that the present invention is provided, mainly
Purpose is to solve to detect the problem of degree of accuracy of SAR image situation of change is relatively low in the prior art.
In order to solve the above problems, present invention generally provides following technical scheme：
In a first aspect, the invention provides a kind of method of detection image change, methods described includes：
Obtain the disparity map of two width image to be detected；
Frequency band parsing is carried out to the disparity map, low frequency sub-band and high-frequency sub-band is obtained；
Linear enhancing processing is carried out to the low frequency sub-band, denoising is carried out to the high-frequency sub-band；
Low frequency sub-band after processing and the high-frequency sub-band after processing are merged, the disparity map after being handled；
Clustering processing is carried out to the disparity map after the processing, the change information between described two width image to be detected is obtained
With non-changing information.
Optionally, described to carry out frequency band parsing to the disparity map, obtaining low frequency sub-band and high-frequency sub-band includes：
Frequency band parsing is carried out to the disparity map by non-lower sampling Shearlet conversion, the low frequency sub-band and institute is obtained
State high-frequency sub-band；
Or, frequency band parsing is carried out to the disparity map by wavelet transformation, the low frequency sub-band and the high frequency is obtained
Subband；
Or, frequency band parsing is carried out to the disparity map by non-downsampling Contourlet conversion, the low frequency is obtained
Subband and the high-frequency sub-band.
Optionally, the high-frequency sub-band after the low frequency sub-band and processing by after processing is merged, after being handled
Disparity map include：
When carrying out frequency band parsing to the disparity map by non-lower sampling Shearlet conversion, pass through non-lower sampling
The inverse transformation of Shearlet conversion is merged to the high-frequency sub-band after the low frequency sub-band after the processing and the processing, is obtained
Obtain the disparity map after the processing；
When carrying out frequency band parsing to the disparity map by wavelet transformation, by the inverse transformation of wavelet transformation to the place
The high-frequency sub-band after low frequency sub-band and the processing after reason is merged, and obtains the disparity map after the processing；
When carrying out frequency band parsing to the disparity map by non-downsampling Contourlet conversion, pass through non-lower sampling
The inverse transformation of contourlet transformation is merged to the high-frequency sub-band after the low frequency sub-band after the processing and the processing,
Obtain the disparity map after the processing.
Optionally, it is described that high-frequency sub-band progress denoising is included：
Suppress the noise in the high-frequency sub-band using Adaptive Thresholding.
Optionally, the disparity map to after the processing, which carries out clustering processing, includes：
Clustering processing is carried out to the disparity map after the processing using fuzzy local message C means clustering algorithms.
Optionally, the disparity map for obtaining two width image to be detected includes：
Smothing filtering denoising is carried out to described two width image to be detected；
Two width image to be detected after denoising are contrasted using neighborhood ratio method is normalized, the disparity map is obtained.
Second aspect, the invention provides a kind of device of detection image change, described device includes：
Acquiring unit, the disparity map for obtaining two width image to be detected；
Resolution unit, the disparity map for being obtained to the acquiring unit carries out frequency band parsing, obtains low frequency sub-band
And high-frequency sub-band；
Processing unit, the low frequency sub-band for being obtained to the resolution unit carries out linear enhancing processing, to described
The high-frequency sub-band that resolution unit is obtained carries out denoising；
Combining unit, for the low frequency sub-band after the processing unit processes and the high-frequency sub-band after processing to be closed
And, the disparity map after being handled；
Cluster cell, clustering processing is carried out for the disparity map after the processing that is obtained to the combining unit, is obtained
Change information and non-changing information between described two width image to be detected.
Optionally, the resolution unit includes：
First parsing module, for carrying out frequency band parsing to the disparity map by non-lower sampling Shearlet conversion, is obtained
Obtain the low frequency sub-band and the high-frequency sub-band；
Second parsing module, for carrying out frequency band parsing to the disparity map by wavelet transformation, obtains low frequency
Band and the high-frequency sub-band；
3rd parsing module, for carrying out frequency band parsing to the disparity map by non-downsampling Contourlet conversion,
Obtain the low frequency sub-band and the high-frequency sub-band.
Optionally, the combining unit includes：
First merging module, for being converted when by non-lower sampling Shearlet to disparity map progress frequency band parsing
When, the inverse transformation converted by non-lower sampling Shearlet is to the high frequency after the low frequency sub-band after the processing and the processing
Subband is merged, and obtains the disparity map after the processing；
Second merging module, for when carrying out frequency band parsing to the disparity map by wavelet transformation, being become by small echo
The inverse transformation changed is merged to the high-frequency sub-band after the low frequency sub-band after the processing and the processing, obtains the processing
Disparity map afterwards；
3rd merging module, frequency band parsing is carried out for working as by non-downsampling Contourlet conversion to the disparity map
When, by the inverse transformation of non-downsampling Contourlet conversion to the height after the low frequency sub-band after the processing and the processing
Frequency subband is merged, and obtains the disparity map after the processing.
Optionally, the processing unit is used to suppress the noise in the high-frequency sub-band using Adaptive Thresholding.
Optionally, the cluster cell is used for using fuzzy local message C means clustering algorithms to the difference after the processing
Different figure carries out clustering processing.
Optionally, the acquiring unit includes：
Denoising module, for carrying out smothing filtering denoising to described two width image to be detected；
Contrast module, neighborhood ratio method is normalized to two width image to be detected after the denoising module denoising for utilizing
Contrasted, obtain the disparity map.
The third aspect, the invention provides a kind of storage medium, the storage medium is stored with a plurality of instruction, the instruction
Method suitable for being loaded by processor and being performed detection image change as described in relation to the first aspect.
Fourth aspect, the invention provides a kind of equipment, the equipment includes：
Processor, is adapted for carrying out each instruction；And
Storage medium, suitable for storing a plurality of instruction；
The instruction is suitable to be loaded by the processor and performed the method for detection image change as described in relation to the first aspect.
By above-mentioned technical proposal, the technical scheme that the present invention is provided at least has following advantages：
Method, device, storage medium and the equipment for the detection image change that the present invention is provided, compared with prior art, no
Only without obtaining priori, and it is not that directly the disparity map is gathered after the disparity map of two images is obtained
Class processing, to obtain image change situation, but first carries out frequency band parsing, acquisition includes effective information most to the differential image
It is many, most important and including the minimum low frequency sub-band of noise, and including the high-frequency sub-band that effective information is minimum and noise is most, so
Linear enhancing processing is carried out to low frequency sub-band afterwards, to strengthen the display intensity of effective information, and high-frequency sub-band is carried out at denoising
Reason, to remove the much noise in high-frequency sub-band, then that the low frequency sub-band and high-frequency sub-band after processing are merged into acquisition is effective
Information enhancement, the disparity map of noise information reduction, finally carry out clustering processing, so as to improve inspection to the disparity map after processing again
The degree of accuracy of altimetric image situation of change.
Described above is only the general introduction of technical solution of the present invention, in order to better understand the technological means of the present invention,
And can be practiced according to the content of specification, and in order to allow above and other objects of the present invention, feature and advantage can
Become apparent, below especially exemplified by the embodiment of the present invention.
Brief description of the drawings
By reading the detailed description of hereafter preferred embodiment, various other advantages and benefit is common for this area
Technical staff will be clear understanding.Accompanying drawing is only used for showing the purpose of preferred embodiment, and is not considered as to the present invention
Limitation.And in view picture accompanying drawing, identical part is denoted by the same reference numerals.In the accompanying drawings：
Fig. 1 shows a kind of flow chart of the method for detection image change provided in an embodiment of the present invention；
Fig. 2 shows a kind of method exemplary plot of detection image change provided in an embodiment of the present invention；
Fig. 3 shows a kind of composition frame chart of the device of detection image change provided in an embodiment of the present invention；
Fig. 4 shows a kind of composition frame chart of the device of detection image change provided in an embodiment of the present invention.
Embodiment
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although showing the disclosure in accompanying drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
Limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
Complete conveys to those skilled in the art.
The embodiments of the invention provide a kind of method of detection image change, as shown in figure 1, this method mainly includes：
101st, the disparity map of two width image to be detected is obtained.
Destination object (i.e. by photography target) is being collected after image not in the same time using equipment such as radars, can be right
Change between any two images is detected.Specifically, because within a period of time, destination object tends not to produce change
Change, so in detection image situation of change, often choosing the two images with intervals and being detected.
After two images to be detected are obtained, directly this two images can be contrasted, to obtain both
Disparity map, but in order that the disparity map that must be obtained is more accurate, can be first with the mode of smothing filtering denoising to this two width
Image carries out preliminary denoising, and to remove additive noise in large quantities, then the image after denoising is contrasted again, obtains difference
Figure.
When obtaining the disparity map of two images, the method used can be differential technique, or ratio method, also may be used
Think other method.Wherein, ratio method includes normalization field ratio method.
102nd, frequency band parsing is carried out to the disparity map, obtains low frequency sub-band and high-frequency sub-band.
Often mixed in piece image by low frequency sub-band and high-frequency sub-band, and frequently included in low frequency sub-band it is a large amount of and
Important effective information, and the noise included in low frequency sub-band is less, and the effective information included in high-frequency sub-band is less (main
For image edge information) and noise is more, therefore after disparity map is obtained, disparity map first can be subjected to frequency band parsing, therefrom
Low frequency sub-band and high-frequency sub-band are obtained, then the feature respectively for low frequency sub-band and high-frequency sub-band is analyzed and handled, i.e.,
Perform step 103.
Wherein, frequency band parsing is carried out to disparity map, so as to obtain the specific algorithm of low frequency sub-band and high-frequency sub-band including non-
Down-sampling Shearlet conversion (NSST conversion), wavelet transformation and non-downsampling Contourlet conversion (NSCT conversion) etc..
That is, the specific implementation of this step can be：Frequency band parsing is carried out to the disparity map by NSST conversion, obtained
The low frequency sub-band and the high-frequency sub-band；
Or, frequency band parsing is carried out to the disparity map by wavelet transformation, the low frequency sub-band and the high frequency is obtained
Subband；
Or, frequency band parsing is carried out to the disparity map by NSCT conversion, the low frequency sub-band and the high frequency is obtained
Subband.
You need to add is that, for the disparity map of SAR image, line frequency is entered to disparity map by the way of NSST is converted
Efficiency with parsing is higher than other modes, therefore, and this step is preferable using NSST transform effects.
103rd, linear enhancing processing is carried out to the low frequency sub-band, denoising is carried out to the high-frequency sub-band.
Referred in above-mentioned steps 102, a large amount of and important effective informations, and low frequency are frequently included in low frequency sub-band
The noise included in band is less, and the effective information included in high-frequency sub-band is less and noise is more, therefore effectively believes in order that obtaining
The display intensity of breath is stronger and as often as possible reduces the noise in image, and linear enhancing processing can be carried out to low frequency sub-band, right
High-frequency sub-band carries out denoising.
Wherein, linear enhanced intensity being carried out to low frequency sub-band can determine according to the feature of image in itself, for example, treat
Image in the low frequency sub-band of processing shows weaker, then linear enhanced intensity can be relatively strong.High-frequency sub-band is gone
Make an uproar the noise predominantly multiplicative noise of processing, and the method for denoising can have a variety of, for example, can be Adaptive Thresholding,
Can be medium filtering, mean filter method etc..
104th, the low frequency sub-band after processing and the high-frequency sub-band after processing are merged, the difference after being handled
Figure.
Linear enhancing processing is being carried out to low frequency sub-band, high-frequency sub-band is being carried out after denoising, can be by after processing
High-frequency sub-band after low frequency sub-band and processing is merged, to obtain the difference of effective information enhancing, noise information reduction
After figure, then carry out Image Change Detection.
When low frequency sub-band and high-frequency sub-band are reconsolidated as a disparity map, the algorithm used is by disparity map
It is decomposed into the inverse operation for the algorithm that low frequency sub-band and high-frequency sub-band are used.That is, being converted when by NSST to the difference
When different figure carries out frequency band parsing, after the inverse transformation converted by NSST is to the low frequency sub-band after the processing and the processing
High-frequency sub-band is merged, and obtains the disparity map after the processing；When by wavelet transformation to the disparity map carry out frequency band solution
During analysis, the high-frequency sub-band after the low frequency sub-band after the processing and the processing is closed by the inverse transformation of wavelet transformation
And, obtain the disparity map after the processing；When carrying out frequency band parsing to the disparity map by NSCT conversion, become by NSCT
The inverse transformation changed is merged to the high-frequency sub-band after the low frequency sub-band after the processing and the processing, obtains the processing
Disparity map afterwards.
105th, clustering processing is carried out to the disparity map after the processing, obtains the change between described two width image to be detected
Information and non-changing information.
, can be using clustering algorithm to the difference after effective information enhancing, the disparity map of noise information reduction is obtained
Pixel in figure carries out clustering processing, therefrom obtains change information between change class and non-changing class, i.e. two images and non-
Change information.Wherein, the clustering algorithm includes but is not limited to following algorithm：Fuzzy local message C mean clusters (FLICM) are calculated
Method, fuzzy C-means clustering (FCM) algorithm and fuzzy c-means (KFCM) algorithm based on kernel function.
You need to add is that, the image that the embodiment of the present invention is detected can be SAR image, or other figures
Picture, its particular type is not limited herein.
The method of detection image change provided in an embodiment of the present invention, compared with prior art, not only without obtaining priori
Knowledge, and be not that clustering processing directly is carried out to the disparity map, to obtain after the disparity map of two images is obtained
Image change situation, but first to the differential image carry out frequency band parsing, acquisition include effective information it is most, most important and including
The minimum low frequency sub-band of noise, and including the high-frequency sub-band that effective information is minimum and noise is most, then low frequency sub-band is entered
Line enhancing is handled, to strengthen the display intensity of effective information, and carries out denoising to high-frequency sub-band, to remove high frequency
Much noise in band, then the low frequency sub-band and high-frequency sub-band after processing are merged into the enhancing of acquisition effective information, noise letter
The reduced disparity map of breath, finally carries out clustering processing, so as to improve detection image situation of change to the disparity map after processing again
The degree of accuracy.
Have normalization neighborhood ratio method using image to be detected by SAR image, using algorithm below, it is NSST conversion, adaptive
Answer exemplified by threshold method, FLICM algorithms, the method that above-mentioned detection image changes is illustrated, as shown in Fig. 2 this method is main
Including：
201st, smothing filtering denoising is carried out to SAR image A, obtains SAR image a；SAR image B is smoothly filtered
Ripple denoising, obtains SAR image b；
202nd, SAR image a and SAR image b are contrasted using normalizing neighborhood ratio method, obtains disparity map M；
203rd, frequency band parsing is carried out to disparity map M by NSST conversion, obtains low frequency sub-band c and high-frequency sub-band d；
204th, linear enhancing processing is carried out to low frequency sub-band c, obtains low frequency sub-band c1；Suppress high by Adaptive Thresholding
Noise in frequency subband d, obtains high-frequency sub-band d1；
205th, the inverse transformation converted by NSST merges processing to low frequency sub-band c1 and high-frequency sub-band d1, obtains difference
Scheme N；
206th, clustering processing is carried out to disparity map N using FLICM algorithms, obtains change class n1 and non-changing class n2.
Wherein, change class n1 and non-changing class n2 is the situation of change between SAR image A and SAR image B.
Further, according to above method embodiment, an alternative embodiment of the invention additionally provides a kind of detection image
The device of change, as shown in figure 3, described device mainly includes：Acquiring unit 31, resolution unit 32, processing unit 33, merging are single
Member 34 and cluster cell 35.Wherein,
Acquiring unit 31, the disparity map for obtaining two width image to be detected；
Resolution unit 32, the disparity map for being obtained to the acquiring unit 31 carries out frequency band parsing, obtains low frequency
Subband and high-frequency sub-band；
Processing unit 33, the low frequency sub-band for being obtained to the resolution unit 32 carries out linear enhancing processing, right
The high-frequency sub-band that the resolution unit 32 is obtained carries out denoising；
Combining unit 34, for the processing unit 33 to be handled after low frequency sub-band and the high-frequency sub-band after processing enter
Row merges, the disparity map after being handled；
Cluster cell 35, clustering processing is carried out for the disparity map after the processing that is obtained to the combining unit 34,
Obtain the change information and non-changing information between described two width image to be detected.
Optionally, as shown in figure 4, the resolution unit 32 includes：
First parsing module 321, for carrying out frequency band parsing to the disparity map by non-lower sampling Shearlet conversion,
Obtain the low frequency sub-band and the high-frequency sub-band；
Second parsing module 322, for carrying out frequency band parsing to the disparity map by wavelet transformation, obtains the low frequency
Subband and the high-frequency sub-band；
3rd parsing module 323, for carrying out frequency band solution to the disparity map by non-downsampling Contourlet conversion
Analysis, obtains the low frequency sub-band and the high-frequency sub-band.
Optionally, as shown in figure 4, the combining unit 34 includes：
First merging module 341, for being converted when by non-lower sampling Shearlet to disparity map progress frequency band solution
During analysis, the inverse transformation converted by non-lower sampling Shearlet is to the height after the low frequency sub-band after the processing and the processing
Frequency subband is merged, and obtains the disparity map after the processing；
Second merging module 342, for when carrying out frequency band parsing to the disparity map by wavelet transformation, passing through small echo
The inverse transformation of conversion is merged to the high-frequency sub-band after the low frequency sub-band after the processing and the processing, obtains the place
Disparity map after reason；
3rd merging module 343, frequency band is carried out for working as by non-downsampling Contourlet conversion to the disparity map
During parsing, by the inverse transformation of non-downsampling Contourlet conversion to the low frequency sub-band after the processing and the processing after
High-frequency sub-band merge, obtain the disparity map after the processing.
Optionally, the processing unit 33 is used to suppress the noise in the high-frequency sub-band using Adaptive Thresholding.
Optionally, after the cluster cell 35 is used for using local message C means clustering algorithms are obscured to the processing
Disparity map carries out clustering processing.
Optionally, as shown in figure 4, the acquiring unit 31 includes：
Denoising module 311, for carrying out smothing filtering denoising to described two width image to be detected；
Contrast module 312, for be checked to two after the denoising of denoising module 311 using neighborhood ratio method is normalized
Altimetric image is contrasted, and obtains the disparity map.
The device of detection image change provided in an embodiment of the present invention, compared with prior art, not only without obtaining priori
Knowledge, and be not that clustering processing directly is carried out to the disparity map, to obtain after the disparity map of two images is obtained
Image change situation, but first to the differential image carry out frequency band parsing, acquisition include effective information it is most, most important and including
The minimum low frequency sub-band of noise, and including the high-frequency sub-band that effective information is minimum and noise is most, then low frequency sub-band is entered
Line enhancing is handled, to strengthen the display intensity of effective information, and carries out denoising to high-frequency sub-band, to remove high frequency
Much noise in band, then the low frequency sub-band and high-frequency sub-band after processing are merged into the enhancing of acquisition effective information, noise letter
The reduced disparity map of breath, finally carries out clustering processing, so as to improve detection image situation of change to the disparity map after processing again
The degree of accuracy.
It should be noted that present apparatus embodiment is corresponding with preceding method embodiment, and for ease of reading, present apparatus embodiment
No longer the detail content in preceding method embodiment is repeated one by one, it should be understood that the device in the present embodiment can
Correspondence realizes the full content in preceding method embodiment.
The device of the detection image change includes processor and memory, and above-mentioned acquiring unit, resolution unit, processing are single
Member, combining unit and cluster cell etc. in memory, storage are stored in by computing device as program unit storage
Said procedure unit in device realizes corresponding function.
Kernel is included in processor, is gone in memory to transfer corresponding program unit by kernel.Kernel can set a width
Or more, the degree of accuracy of detection SAR image situation of change is improved by adjusting kernel parameter.
Memory potentially includes the volatile memory in computer-readable medium, random access memory (RAM) and/
Or the form, such as read-only storage (ROM) or flash memory (flashRAM) such as Nonvolatile memory, memory includes the storage of an at least width
Chip.
The embodiments of the invention provide a kind of storage medium, the storage medium is stored with a plurality of instruction, and the instruction is suitable
Method for being loaded by processor and being performed above-mentioned detection image change.
The storage medium of detection image change provided in an embodiment of the present invention, compared with prior art, not only without obtaining
Priori, and be not that clustering processing directly is carried out to the disparity map after the disparity map of two images is obtained, so as to
Obtain image change situation, but first to the differential image carry out frequency band parsing, acquisition include effective information it is most, most important and
Including the minimum low frequency sub-band of noise, and including the high-frequency sub-band that effective information is minimum and noise is most, then to low frequency
Bring line enhancing processing into, to strengthen the display intensity of effective information, and denoising is carried out to high-frequency sub-band, to remove height
Much noise in frequency subband, then by the low frequency sub-band and high-frequency sub-band after processing merge acquisition effective information enhancing, make an uproar
The disparity map of acoustic intelligence reduction, finally carries out clustering processing to the disparity map after processing again, so as to improve detection image change
The degree of accuracy of situation.
The embodiments of the invention provide a kind of processor, the processor is used for operation program, wherein, described program operation
The method of the above-mentioned detection image changes of Shi Zhihang.
The embodiments of the invention provide a kind of equipment, equipment includes processor, storage medium.Wherein, processor, suitable for reality
Now each instruction；And storage medium, suitable for storing a plurality of instruction；The instruction is suitable to be loaded and performed by the processor：
Obtain the disparity map of two width image to be detected；
Frequency band parsing is carried out to the disparity map, low frequency sub-band and high-frequency sub-band is obtained；
Linear enhancing processing is carried out to the low frequency sub-band, denoising is carried out to the high-frequency sub-band；
Low frequency sub-band after processing and the high-frequency sub-band after processing are merged, the disparity map after being handled；
Clustering processing is carried out to the disparity map after the processing, the change information between described two width image to be detected is obtained
With non-changing information.
Optionally, described to carry out frequency band parsing to the disparity map, obtaining low frequency sub-band and high-frequency sub-band includes：
Frequency band parsing is carried out to the disparity map by non-lower sampling Shearlet conversion, the low frequency sub-band and institute is obtained
State high-frequency sub-band；
Or, frequency band parsing is carried out to the disparity map by wavelet transformation, the low frequency sub-band and the high frequency is obtained
Subband；
Or, frequency band parsing is carried out to the disparity map by non-downsampling Contourlet conversion, the low frequency is obtained
Subband and the high-frequency sub-band.
Optionally, the high-frequency sub-band after the low frequency sub-band and processing by after processing is merged, after being handled
Disparity map include：
When carrying out frequency band parsing to the disparity map by non-lower sampling Shearlet conversion, pass through non-lower sampling
The inverse transformation of Shearlet conversion is merged to the high-frequency sub-band after the low frequency sub-band after the processing and the processing, is obtained
Obtain the disparity map after the processing；
When carrying out frequency band parsing to the disparity map by wavelet transformation, by the inverse transformation of wavelet transformation to the place
The high-frequency sub-band after low frequency sub-band and the processing after reason is merged, and obtains the disparity map after the processing；
When carrying out frequency band parsing to the disparity map by non-downsampling Contourlet conversion, pass through non-lower sampling
The inverse transformation of contourlet transformation is merged to the high-frequency sub-band after the low frequency sub-band after the processing and the processing,
Obtain the disparity map after the processing.
Optionally, it is described that high-frequency sub-band progress denoising is included：
Suppress the noise in the high-frequency sub-band using Adaptive Thresholding.
Optionally, the disparity map to after the processing, which carries out clustering processing, includes：
Clustering processing is carried out to the disparity map after the processing using fuzzy local message C means clustering algorithms.
Optionally, the disparity map for obtaining two width image to be detected includes：
Smothing filtering denoising is carried out to described two width image to be detected；
Two width image to be detected after denoising are contrasted using neighborhood ratio method is normalized, the disparity map is obtained.
It should be noted that equipment herein can be server, PC, PAD, mobile phone etc..
The equipment of detection image change provided in an embodiment of the present invention, compared with prior art, not only without obtaining priori
Knowledge, and be not that clustering processing directly is carried out to the disparity map, to obtain after the disparity map of two images is obtained
Image change situation, but first to the differential image carry out frequency band parsing, acquisition include effective information it is most, most important and including
The minimum low frequency sub-band of noise, and including the high-frequency sub-band that effective information is minimum and noise is most, then low frequency sub-band is entered
Line enhancing is handled, to strengthen the display intensity of effective information, and carries out denoising to high-frequency sub-band, to remove high frequency
Much noise in band, then the low frequency sub-band and high-frequency sub-band after processing are merged into the enhancing of acquisition effective information, noise letter
The reduced disparity map of breath, finally carries out clustering processing, so as to improve detection image situation of change to the disparity map after processing again
The degree of accuracy.
Present invention also provides a kind of computer program product, when being performed on data processing equipment, it is adapted for carrying out just
The program code of beginningization there are as below methods step：
Obtain the disparity map of two width image to be detected；
Frequency band parsing is carried out to the disparity map, low frequency sub-band and high-frequency sub-band is obtained；
Linear enhancing processing is carried out to the low frequency sub-band, denoising is carried out to the high-frequency sub-band；
Low frequency sub-band after processing and the high-frequency sub-band after processing are merged, the disparity map after being handled；
Clustering processing is carried out to the disparity map after the processing, the change information between described two width image to be detected is obtained
With non-changing information.
It should be understood by those skilled in the art that, embodiments herein can be provided as method, system or computer program
Product.Therefore, the application can be using the reality in terms of complete hardware embodiment, complete software embodiment or combination software and hardware
Apply the form of example.Moreover, the application can be using the computer for wherein including computer usable program code in one or more
The computer program production that usable storage medium is implemented on (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The application is the flow with reference to method, equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram are described.It should be understood that can be by every first-class in computer program instructions implementation process figure and/or block diagram
Journey and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computer programs can be provided
The processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw width machine so that produced by the instruction of computer or the computing device of other programmable data processing devices for real
The device for the function of being specified in the present width flow of flow chart one or several flows and/or the width square frame of block diagram one or several square frames.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which is produced, to be included referring to
Make the manufacture of device, the command device realize in the width flow of flow chart one or several flows and/or the width square frame of block diagram one or
The function of being specified in several square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that in meter
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, thus in computer or
The instruction performed on other programmable devices is provided for realizing in the width flow of flow chart one or several flows and/or block diagram one
The step of function of being specified in width square frame or several square frames.
In a width typically configuration, computing device includes one or more processor (CPU), input/output interface, net
Network interface and internal memory.
Memory potentially includes the volatile memory in computer-readable medium, random access memory (RAM) and/
Or the form, such as read-only storage (ROM) or flash memory (flashRAM) such as Nonvolatile memory.Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer-readable instruction, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moved
State random access memory (DRAM), other kinds of random access memory (RAM), read-only storage (ROM), electric erasable
Programmable read only memory (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc read-only storage (CD-ROM),
Digital versatile disc (DVD) or other optical storages, magnetic cassette tape, the storage of tape magnetic rigid disk or other magnetic storage apparatus
Or any other non-transmission medium, the information that can be accessed by a computing device available for storage.Define, calculate according to herein
Machine computer-readable recording medium does not include temporary computer readable media (transitory media), such as data-signal and carrier wave of modulation.
It should also be noted that, term " comprising ", "comprising" or its any other variant are intended to nonexcludability
Comprising so that process, method, commodity or equipment including a series of key elements are not only including those key elements, but also wrap
Include other key elements being not expressly set out, or also include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the key element limited by sentence " including a width ... ", it is not excluded that including key element
Also there is other identical element in process, method, commodity or equipment.
It will be understood by those skilled in the art that embodiments herein can be provided as method, system or computer program product.
Therefore, the application can be using the embodiment in terms of complete hardware embodiment, complete software embodiment or combination software and hardware
Form.Deposited moreover, the application can use the computer for wherein including computer usable program code in one or more can use
The shape for the computer program product that storage media is implemented on (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
Embodiments herein is these are only, the application is not limited to.To those skilled in the art,
The application can have various modifications and variations.All any modifications made within spirit herein and principle, equivalent substitution,
Improve etc., it should be included within the scope of claims hereof.
Claims (10)
1. a kind of method of detection image change, it is characterised in that methods described includes：
Obtain the disparity map of two width image to be detected；
Frequency band parsing is carried out to the disparity map, low frequency sub-band and high-frequency sub-band is obtained；
Linear enhancing processing is carried out to the low frequency sub-band, denoising is carried out to the high-frequency sub-band；
Low frequency sub-band after processing and the high-frequency sub-band after processing are merged, the disparity map after being handled；
Clustering processing is carried out to the disparity map after the processing, change information between described two width image to be detected is obtained and non-
Change information.
2. according to the method described in claim 1, it is characterised in that described that frequency band parsing is carried out to the disparity map, obtain low
Frequency subband and high-frequency sub-band include：
Frequency band parsing is carried out to the disparity map by non-lower sampling Shearlet conversion, the low frequency sub-band and the height is obtained
Frequency subband；
Or, frequency band parsing is carried out to the disparity map by wavelet transformation, the low frequency sub-band and the high-frequency sub-band is obtained；
Or, frequency band parsing is carried out to the disparity map by non-downsampling Contourlet conversion, the low frequency sub-band is obtained
With the high-frequency sub-band.
3. method according to claim 2, it is characterised in that the height after the low frequency sub-band and processing by after processing
Frequency subband is merged, and the disparity map after being handled includes：
When carrying out frequency band parsing to the disparity map by non-lower sampling Shearlet conversion, pass through non-lower sampling Shearlet
The inverse transformation of conversion is merged to the high-frequency sub-band after the low frequency sub-band after the processing and the processing, obtains the place
Disparity map after reason；
When by wavelet transformation to the disparity map carry out frequency band parsing when, by the inverse transformation of wavelet transformation to the processing after
Low frequency sub-band and the processing after high-frequency sub-band merge, obtain the disparity map after the processing；
When carrying out frequency band parsing to the disparity map by non-downsampling Contourlet conversion, pass through non-lower sampling
The inverse transformation of contourlet transformation is merged to the high-frequency sub-band after the low frequency sub-band after the processing and the processing,
Obtain the disparity map after the processing.
4. according to the method described in claim 1, it is characterised in that described that high-frequency sub-band progress denoising is included：
Suppress the noise in the high-frequency sub-band using Adaptive Thresholding.
5. method according to any one of claim 1 to 4, it is characterised in that the disparity map to after the processing
Carrying out clustering processing includes：
Clustering processing is carried out to the disparity map after the processing using fuzzy local message C means clustering algorithms.
6. method according to any one of claim 1 to 4, it is characterised in that described width image to be detected of acquisition two
Disparity map includes：
Smothing filtering denoising is carried out to described two width image to be detected；
Two width image to be detected after denoising are contrasted using neighborhood ratio method is normalized, the disparity map is obtained.
7. a kind of device of detection image change, it is characterised in that described device includes：
Acquiring unit, the disparity map for obtaining two width image to be detected；
Resolution unit, the disparity map for being obtained to the acquiring unit carries out frequency band parsing, obtains low frequency sub-band and height
Frequency subband；
Processing unit, the low frequency sub-band for being obtained to the resolution unit carries out linear enhancing processing, to the parsing
The high-frequency sub-band that unit is obtained carries out denoising；
Combining unit, for the low frequency sub-band after the processing unit processes and the high-frequency sub-band after processing to be merged,
Disparity map after being handled；
Cluster cell, clustering processing is carried out for the disparity map after the processing that is obtained to the combining unit, obtains described
Change information and non-changing information between two width image to be detected.
8. device according to claim 7, it is characterised in that the resolution unit includes：
First parsing module, for carrying out frequency band parsing to the disparity map by non-lower sampling Shearlet conversion, obtains institute
State low frequency sub-band and the high-frequency sub-band；
Second parsing module, for carrying out frequency band parsing to the disparity map by wavelet transformation, obtain the low frequency sub-band and
The high-frequency sub-band；
3rd parsing module, for carrying out frequency band parsing to the disparity map by non-downsampling Contourlet conversion, is obtained
The low frequency sub-band and the high-frequency sub-band.
9. a kind of storage medium, it is characterised in that the storage medium is stored with a plurality of instruction, the instruction is applied to by handling
The method that device loads and performs the change of the detection image as described in any one in claim 1 to claim 6.
10. a kind of equipment, it is characterised in that the equipment includes：
Processor, is adapted for carrying out each instruction；And
Storage medium, suitable for storing a plurality of instruction；
The instruction is suitable to be loaded and performed as described in any one in claim 1 to claim 6 as the processor
The method of detection image change.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710393356.XA CN107103595A (en) | 2017-05-27 | 2017-05-27 | Method, device, storage medium and the equipment of detection image change |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710393356.XA CN107103595A (en) | 2017-05-27 | 2017-05-27 | Method, device, storage medium and the equipment of detection image change |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107103595A true CN107103595A (en) | 2017-08-29 |
Family
ID=59659265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710393356.XA Pending CN107103595A (en) | 2017-05-27 | 2017-05-27 | Method, device, storage medium and the equipment of detection image change |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107103595A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107590788A (en) * | 2017-09-12 | 2018-01-16 | 新疆大学 | A kind of remote sensing image processing method |
| CN107862666A (en) * | 2017-11-22 | 2018-03-30 | 新疆大学 | Mixing Enhancement Methods about Satellite Images based on NSST domains |
| CN107945195A (en) * | 2017-11-10 | 2018-04-20 | 西安电子科技大学 | The Polarimetric SAR Image change detecting method of DBN is integrated based on profile ripple |
| CN108932492A (en) * | 2018-06-28 | 2018-12-04 | 福州昌宇五金锁具制品有限公司 | A kind of image fingerprint extracting method based on non-sampled shearing wave conversion |
| CN109978855A (en) * | 2019-03-26 | 2019-07-05 | 新疆大学 | A kind of method for detecting change of remote sensing image and device |
| CN112308025A (en) * | 2020-11-23 | 2021-02-02 | 内蒙古工业大学 | Image change recognition method and device, storage medium and electronic equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008229864A (en) * | 2007-03-16 | 2008-10-02 | Canon Inc | Inkjet recording method and image processing device |
| CN106384339A (en) * | 2016-09-30 | 2017-02-08 | 防城港市港口区高创信息技术有限公司 | Infrared night vision image enhancement method |
-
2017
- 2017-05-27 CN CN201710393356.XA patent/CN107103595A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008229864A (en) * | 2007-03-16 | 2008-10-02 | Canon Inc | Inkjet recording method and image processing device |
| CN106384339A (en) * | 2016-09-30 | 2017-02-08 | 防城港市港口区高创信息技术有限公司 | Infrared night vision image enhancement method |
Non-Patent Citations (3)
| Title |
|---|
| 张一晨等: "基于NSCT域图像融合改进算法的SAR图像无监督变化检测", 《光电子激光》 * |
| 李玲玲等: "混合的SAR图像变化检测算法", 《计算机工程与设计》 * |
| 李青松等: "基于非下采样Shearlet和几何结构的遥感图像无监督变化检测", 《遥感技术与应用》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107590788A (en) * | 2017-09-12 | 2018-01-16 | 新疆大学 | A kind of remote sensing image processing method |
| CN107945195A (en) * | 2017-11-10 | 2018-04-20 | 西安电子科技大学 | The Polarimetric SAR Image change detecting method of DBN is integrated based on profile ripple |
| CN107945195B (en) * | 2017-11-10 | 2019-10-25 | 西安电子科技大学 | The polarimetric SAR image change detecting method of DBN is integrated based on profile wave |
| CN107862666A (en) * | 2017-11-22 | 2018-03-30 | 新疆大学 | Mixing Enhancement Methods about Satellite Images based on NSST domains |
| CN108932492A (en) * | 2018-06-28 | 2018-12-04 | 福州昌宇五金锁具制品有限公司 | A kind of image fingerprint extracting method based on non-sampled shearing wave conversion |
| CN109978855A (en) * | 2019-03-26 | 2019-07-05 | 新疆大学 | A kind of method for detecting change of remote sensing image and device |
| CN112308025A (en) * | 2020-11-23 | 2021-02-02 | 内蒙古工业大学 | Image change recognition method and device, storage medium and electronic equipment |
| CN112308025B (en) * | 2020-11-23 | 2023-01-31 | 内蒙古工业大学 | Image change identification method and device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107103595A (en) | Method, device, storage medium and the equipment of detection image change | |
| Marchesi et al. | A context-sensitive technique robust to registration noise for change detection in VHR multispectral images | |
| Kang et al. | Robust median filtering forensics using an autoregressive model | |
| Qu et al. | A pedestrian detection method based on yolov3 model and image enhanced by retinex | |
| CN111709883A (en) | Image detection method, device and equipment | |
| CN107704797B (en) | Real-time detection method, system and equipment based on pedestrians and vehicles in security video | |
| CN106570862A (en) | Super-resolution reconstruction quality evaluation method and apparatus thereof | |
| Raj et al. | Medical image denoising using multi-resolution transforms | |
| CN107301641A (en) | A kind of detection method and device of Remote Sensing Imagery Change | |
| Wu et al. | Superpixel-based image noise variance estimation with local statistical assessment | |
| CN108710851B (en) | Seismic signal random noise attenuation method, terminal device and storage medium | |
| Chen et al. | Two-pass bilateral smooth filtering for remote sensing imagery | |
| Denton et al. | Multi-scalar analysis of geospatial agricultural data for sustainability | |
| Morandeira et al. | Assessment of SAR speckle filters in the context of object-based image analysis | |
| CN107316296A (en) | A kind of method for detecting change of remote sensing image and device based on logarithmic transformation | |
| CN104200438A (en) | Multi-level infrared image detail enhancement processing method and processing device thereof | |
| CN116309612B (en) | Semiconductor silicon wafer detection method, device and medium based on frequency decoupling supervision | |
| HosseinKhani et al. | Real‐time removal of impulse noise from MR images for radiosurgery applications | |
| Tumakov et al. | Elimination of defects in mammograms caused by a malfunction of the device matrix | |
| CN114463764A (en) | Table line detection method and device, computer equipment and storage medium | |
| Chen et al. | Structural damage detection using bi-temporal optical satellite images | |
| Lyu | Research on subway pedestrian detection algorithm based on big data cleaning technology | |
| AKINTOYE et al. | COMPOSITE MEDIAN WIENER FILTER BASED TECHNIQUE FOR IMAGE ENHANCEMENT. | |
| Ezekiel et al. | No-reference blur metric using double-density and dual-tree two-dimensional wavelet transformation | |
| Gupta et al. | Object recognition based on template matching and correlation method in hyperspectral images |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170829 |
|
| RJ01 | Rejection of invention patent application after publication |