CN106251288B - Twin-line array image device subpixel image synthetic method based on Multi-resolution analysis - Google Patents
Twin-line array image device subpixel image synthetic method based on Multi-resolution analysis Download PDFInfo
- Publication number
- CN106251288B CN106251288B CN201610579045.8A CN201610579045A CN106251288B CN 106251288 B CN106251288 B CN 106251288B CN 201610579045 A CN201610579045 A CN 201610579045A CN 106251288 B CN106251288 B CN 106251288B
- Authority
- CN
- China
- Prior art keywords
- image
- resolution
- filter
- pass
- low
- 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.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 32
- 238000010189 synthetic method Methods 0.000 title claims abstract description 24
- 238000000034 method Methods 0.000 claims abstract description 46
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000013461 design Methods 0.000 claims description 25
- 238000003384 imaging method Methods 0.000 claims description 19
- 239000011159 matrix material Substances 0.000 claims description 15
- 238000001914 filtration Methods 0.000 claims description 13
- 230000009466 transformation Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 11
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 7
- 230000003287 optical effect Effects 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 230000004044 response Effects 0.000 claims description 3
- 230000035939 shock Effects 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 abstract description 9
- 230000015572 biosynthetic process Effects 0.000 abstract description 6
- 238000003786 synthesis reaction Methods 0.000 abstract description 6
- 230000002950 deficient Effects 0.000 abstract description 4
- 230000004304 visual acuity Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 239000000463 material Substances 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 230000008878 coupling Effects 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 4
- 238000005859 coupling reaction Methods 0.000 description 4
- 238000003491 array Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 210000001747 pupil Anatomy 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
This application discloses a kind of twin-line array image device subpixel image synthetic methods based on Multi-resolution analysis, respectively be staggered on CCD linear array direction and heading half of pixel sampling respectively, to obtain the image of two width low resolutions, respectively image A and image B.Using image B as reference picture, to image A Histogram Matching, image C after being matched, then after image B and image C is carried out image registration, image B and image C is synthesized into a width and combines image.Interpolation processing is carried out to combination image, the image Jing Guo interpolation processing is finally filtered reconstruction by the filter of special designing, and then obtain the image of high resolution.Such scheme, in the case where not changing system parameter, using the method for the over-sampling Data Synthesis high resolution image of acquisition, with solve the problems, such as at present due to caused by high resolution CCD is deficient picture resolution it is limited, the effective utilization of system is greatly improved, cost is reduced.
Description
Technical field
This application involves spatial remotely sensed imaging technical fields, specifically, being related to a kind of based on the double of Multi-resolution analysis
Linear array imaging device subpixel image synthetic method.
Background technique
In the industrial circles such as aerospace, the chemical industry energy, many important components such as turbogenerator, supersonic vehicle
Surface texture etc. all can long service in the high temperature environment.But the research method since hot environment is very severe, under room temperature
It is difficult to apply to high temperature, therefore currently about the various researchs of material under high temperature performance change mostly all in qualitative research
Stage, if can to material, material properties change procedure is quantitatively accurately observed and is measured in a high temperauture environment, energy
The preferably different material behavior features with room temperature at a high temperature of further investigation material are high so as to break through these problems bring
Warm barrier.
It is one of most important developing goal of imaging field that system resolution, which is improved, to obtain HD image always.Mesh
Before, electro-optical system is generally used CCD (Charge Coupled Device) array device and is imaged.Influence this typical case
Optics-discrete sampling system imaging quality factor mainly from two aspect: optical imaging system process, image recording and
Digitized process.Start with from these two aspects, the method for improving digital imaging system resolving power mainly has pure optical method, including so-called
Micro lens technology (on chip μ-lens) and Pupil Super-Resolution power technology etc.；Pure Hardware Method improves the production work of CCD
Skill reduces pixel area, image recovery method etc..
Sub-pixel imaging technique is in the case where not changing optical system parameter, by improving imaging system detector
Room and time sample frequency be starting point, using mechanical scanning mode (to line array CCD) or minute movement (micro scanning)
The mode of software post-processing synthesis, a kind of very effective method of Lai Tigao picture resolution are recycled after CCD more times imagings.
It according to the difference of research contents and implementation, can be classified as: realize sub-pixel imaging technique using spatial information, utilize
Temporal information realizes sub-pixel imaging technique, realizes sub-pixel imaging technique using spectral information.Sub-pixel imaging technique can be with
The digital picture that high resolution is obtained using existing low resolution CCD increases substantially the effective utilization of system, reduces cost.
Due to national defence and Military Application, high resolution CCD is always the device of export restrictions in the world.Therefore, sub_pixel model
Technology has great importance to the resolving power for improving the fields high request images such as aerospace and practicability, and being one is worth closing
The research direction of note and long-run development.
The key step of sub_pixel model process under normal circumstances are as follows:
1, two rows of (to linear arrays) or accurate half of pixel dislocation of four rows (opposite battle array) CCD are placed, obtains low resolution figure
Picture；
2, using the interpolation for being based on Shannon (Shannon) sampling thheorem, pixel value is encrypted with high confidence level；
3, image restoration is carried out, high resolution image is obtained.
Currently, the sub-pixed mapping composition algorithm overwhelming majority is the algorithm based on linear interpolation thought, the common spy of these algorithms
Point is some pixel points only considered around required pixel point, without effectively being analyzed entire image, leads to image point
It is limited to distinguish power, and the process that image restoration is carried out in step 3 is extremely complex.
Summary of the invention
In view of this, the technical problem to be solved by the application is to provide a kind of twin-line arrays based on Multi-resolution analysis
Image device subpixel image synthetic method provides for twin-line array CCD sub-pixel imaging technique, is not changing system parameter
In the case of, using the method for the over-sampling Data Synthesis high resolution image of acquisition, to solve at present since high resolution CCD is deficient
The limited problem of picture resolution caused by weary, greatly improves the effective utilization of system, reduces cost.
In order to solve the above-mentioned technical problem, the application has following technical solution:
A kind of twin-line array image device subpixel image synthetic method based on Multi-resolution analysis, which is characterized in that packet
It includes:
In the imaging detector using line array CCD as representative, both threads battle array CCD is encapsulated in same focal plane device,
Or both threads battle array is imaged simultaneously by the means of optical spectroscopic, by both threads battle array CCD in linear array direction and each mistake of heading
Half of pixel is opened, obtains two width low resolution image A and image B respectively；
Using described image B as reference picture, histogram transformation is carried out to described image A, so that the histogram of described image A
It is consistent with the histogram of described image B, the image C after being matched；
Image registration is carried out to described image B and described image C；
Using Multi-resolution analyze in five plants of arrangement modes, by after image registration image B and image C synthesize
One width combines image；
Interpolation processing is carried out to the combination image, by vacant grid outside dislocation two rows of on the combination image position CCD
Point zero padding；
Design low pass resolution filter H0, high pass resolution filter H1, low-pass reconstruction filters G0With high pass reconfigurable filter
G1；
To the combination image after the zero padding, the low pass resolution filter H of design is utilized0, high pass resolution filter H1, it is low
Logical reconfigurable filter G0With high pass reconfigurable filter G1It is decomposed and is rebuild, obtain high resolution image.
Preferably, in which:
Both threads battle array CCD is respectively staggered half of pixel in linear array direction and heading, obtains two width low resolutions respectively
Image A and image B, further are as follows:
Using the sub_pixel model means of prismatic decomposition technology, in the linear array direction of ccd detector and each mistake of heading
It opens half of pixel to be sampled, according to the linear array direction and the heading, the ccd detector centre coordinate is distinguished
It is set as (0,0) and (0, p/2), by push-scanning image, both threads battle array CCD gets two width low resolution image A and image simultaneously
B。
Preferably, in which:
It is described that image registration is carried out to described image B and image C, further are as follows:
By being based on image First-order Gradient, calculate the sub-pixed mapping of variance and minimum registration between the image by linear interpolation
Method carries out image registration to described image B and described image C, specifically:
Based on gradient G (Gx, Gy) formula are as follows:
Gx, GyGradient value on the direction x, y is respectively indicated, summation symbol is to i and j, and (i, j) is pixel position, in which:
For image A, have:
A [i, j] represents the pixel value at (i, j),The average pixel value at (i, j) is represented,Represent (i,
J) direction the x average gradient value at place,The average gradient value in the direction y at (i, j) is represented,
Simultaneous following equations eliminate α1、β1、α2、β2, obtain the dx, dy compensated by deviation from linearity:
Two groups of equations can be by translating A image S in the direction x, y respectively afterwardsx、SyAfter a unit, solved with B image registration
Obtain (Sx、SyIt is taken as ± 1).
Preferably, in which:
The low pass resolution filter H of design0, the low-pass reconstruction filters G0, the high pass resolution filter H1With
And the high pass reconfigurable filter G1For the filter of the nondecomposable five plant shapes particular form of two dimension.
Preferably, in which:
The low pass resolution filter H0With the low-pass reconstruction filters G0It designs by the following method:
One-dimensional filtering device is first expanded to two dimensional filter by separable method, then again by two by utilization level rotary process
It ties up filter and rotates 45 °, obtain new filter, wherein H0、H1、G0And G1Meet:
H0(z1,z2)H1(-z1,-z2)-H0(-z1,-z2)H1(z1,z2)=const
Using alternative transforms method, the construction two-dimensional linear phase filter since One Dimensional Polynomial matrix, the one of selection
Tie up polynomial matrix are as follows:
According to linear phase reconstruction filtering theorem, enable:
C=ad/a1
Obtain the shock response of filter are as follows:
Take a=1, a1=-4, a2=28, d=-4 are utilizedRelationship, obtain low pass
Resolution filter H0With low-pass reconstruction filters G0Matrix be respectively as follows:
Wherein,Respectively indicate 4 piecemeal squares of resolution filter
Battle array, a, d are constants.
Preferably, in which:
The high pass resolution filter H1With the high pass reconfigurable filter G1It designs by the following method:
Utilize formula:
According to above-mentioned formula, in conjunction with the low pass resolution filter H0With the low-pass reconstruction filters G0Design method
It obtains.
Preferably, in which:
Interpolation processing is carried out to the combination image, by vacant grid outside dislocation two rows of on the combination image position CCD
After point zero padding, the temporal relationship of respective coordinates point data are as follows:
Wherein, (n1,n2) representative image grid point, the position of x, y representation transformation front and back each grid point of image.To y (n1,
n2) carry out transform can obtain:
It enables in above formulaObtain the frequency domain relationship of respective coordinates point data are as follows:
Preferably, in which:
To the combination image after the zero padding, the low pass resolution filter H of design is utilized0, high pass resolution filter H1, it is low
Logical reconfigurable filter G0With high pass reconfigurable filter G1It is decomposed and is rebuild, obtain high resolution image, further are as follows:
Utilize the low pass resolution filter H of design0, high pass resolution filter H1, low-pass reconstruction filters G0It is reconstructed with high pass
Filter G1, in the low-pass reconstruction filters G0Matrix defined by within the scope of airspace, to the combination image after the zero padding
Low-pass filtering treatment and reconstruction are carried out, high resolution image is obtained.
Compared with prior art, method described herein achieving the following effects:
First, the twin-line array image device subpixel image synthetic method of the present invention based on Multi-resolution analysis, from
The angle analysis integrally filtered the imaging problem of sub-pixed mapping, to entire image has all carried out effective analysis, theoretically can will point
It distinguishes that power rises to 1.42 times of original image, there is higher raising resolving power ability.
Second, the twin-line array image device subpixel image synthetic method of the present invention based on Multi-resolution analysis, benefit
, can also be with accuracy registration for noisy acoustic image with the information of entire image, therefore have the resolving power of noise image to raising
With better effect.
Third, the twin-line array image device subpixel image synthetic method of the present invention based on Multi-resolution analysis,
It is current to solve using the method for the over-sampling Data Synthesis high resolution image of acquisition in the case where not changing system parameter
Due to high resolution CCD it is deficient caused by the limited problem of picture resolution, greatly improve the effective utilization of system, reduce at
This.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen
Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:
Fig. 1 is the stream of the twin-line array image device subpixel image synthetic method of the present invention based on Multi-resolution analysis
Cheng Tu；
Fig. 2 is the Amici prism schematic diagram in the present invention；
Fig. 3 is the flow diagram of Histogram Matching in the present invention；
Fig. 4 is that dislocation five ancient unit of weight, equal to 1/24 liang of CCD obtains image front and rear coordinate schematic diagram in the present invention；
Fig. 5 is that sub_pixel model of the present invention dislocation obtains, zero padding and filtering reconstruct schematic diagram；
Fig. 6 is five plant shape zero padding front and rear coordinate schematic diagrames in the present invention；
Fig. 7 is the twin-line array image device subpixel image synthesis in Application Example of the present invention based on Multi-resolution analysis
The flow chart of method.
Specific embodiment
As used some vocabulary to censure specific components in the specification and claims.Those skilled in the art answer
It is understood that hardware manufacturer may call the same component with different nouns.This specification and claims are not with name
The difference of title is as the mode for distinguishing component, but with the difference of component functionally as the criterion of differentiation.Such as logical
The "comprising" of piece specification and claim mentioned in is an open language, therefore should be construed to " include but do not limit
In "." substantially " refer within the acceptable error range, those skilled in the art can within a certain error range solve described in
Technical problem basically reaches the technical effect.In addition, " coupling " word includes any direct and indirect electric property coupling herein
Means.Therefore, if it is described herein that a first device is coupled to a second device, then representing the first device can directly electrical coupling
It is connected to the second device, or the second device indirectly electrically coupled through other devices or coupling means.Specification
Subsequent descriptions be implement the application better embodiment, so it is described description be for the purpose of the rule for illustrating the application,
It is not intended to limit the scope of the present application.The protection scope of the application is as defined by the appended claims.
Embodiment 1
Shown in Figure 1 is a kind of herein described twin-line array image device subpixel image based on Multi-resolution analysis
The flow chart of synthetic method, this include: based on the twin-line array image device subpixel image synthetic method that Multi-resolution is analyzed
Step 101, in the imaging detector using line array CCD as representative, both threads battle array CCD is encapsulated in same focal plane
In device, or by the means of optical spectroscopic both threads battle array is imaged simultaneously, by both threads battle array CCD in linear array direction and flight side
To half of the pixel that is respectively staggered, two width low resolution image A and image B are obtained respectively；
Step 102, using described image B as reference picture, to described image A carry out histogram transformation so that described image A
Histogram it is consistent with the histogram of described image B, the image C after being matched；
Step 103 carries out image registration to described image B and image C；
Step 104, using Multi-resolution analyze in five plants of arrangement modes, by the image B and figure after image registration
As C synthesizes width combination image；
Step 105 carries out interpolation processing to the combination image, will be outside dislocation two rows of on the combination image position CCD
Vacant grid point zero padding；
Step 106, design low pass resolution filter H0, high pass resolution filter H1, low-pass reconstruction filters G0With high pass weight
Structure filter G1；
Step 107, to the combination image after the zero padding, utilize the low pass resolution filter H of design0, high pass decompose filter
Wave device H1, low-pass reconstruction filters G0With high pass reconfigurable filter G1It is decomposed and is rebuild, obtain high resolution image.
The above-mentioned twin-line array image device subpixel image synthetic method based on Multi-resolution analysis of the invention, obtains figure
As during, be respectively staggered on CCD linear array direction and heading half of pixel sampling respectively, to obtain two low resolutions
The image of power, respectively image A and image B.Using image B as reference picture, to image A Histogram Matching, after being matched
Image C, then by after image B and image C progress image registration, image B and image C is synthesized into a width and combines image.To constitutional diagram
As carrying out interpolation processing, the image Jing Guo interpolation processing is finally filtered reconstruction by the filter of special designing, in turn
Obtain the image of high resolution.
This method can effectively divide entire image from the imaging problem of the angle analysis sub-pixed mapping integrally filtered
Analysis, in this manner, being more advantageous to compared with the simple algorithm for considering some pixel points around required pixel point in the prior art
Improve the resolving power of image.And the above method provided by the present invention can also influence effectively to be analyzed on picture noise, and it is right
Improving has the resolving power of noise image to have better effect.
Embodiment 2
On the basis of 1 the method for embodiment, above-mentioned steps 101, by both threads battle array CCD in linear array direction and flight side
To half of the pixel that is respectively staggered, two width low resolution image A and image B are obtained respectively, further are as follows:
Using prismatic decomposition technology, in the linear array direction of ccd detector and heading, half of the pixel that is respectively staggered is adopted
The ccd detector according to the linear array direction and the heading, (is presented as CCD referring to fig. 2 by sample in figureAWith
CCDB) centre coordinate is set to (0,0) and (0, p/2), during push-scanning image, two CCD get two low point simultaneously
It distinguishes and tries hard to as A and image B.
In above-mentioned steps 102, the process of Histogram Matching passes through detector C CD referring to Fig. 3AImage A is got, is passed through
Detector C CDBImage B is got, using image B as reference picture, histogram transformation is carried out to image A, so that image A's is straight
Side's figure is consistent with the histogram of image B, thus the new images C after being matched.In later interpolation processing, by new images
(i.e. image C) replaces the image (i.e. image A) before transformation.
In above-mentioned steps 103, image registration is carried out to described image B and image C, further are as follows:
By being based on image First-order Gradient, calculate the sub-pixed mapping of variance and minimum registration between the image by linear interpolation
Method carries out image registration to described image B and described image C, which is utilized the information of whole image, for noisy
Acoustic image can also with accuracy registration, specifically:
Based on gradient G (Gx, Gy) formula are as follows:
Gx, GyGradient value on the direction x, y is respectively indicated, summation symbol is to i and j, and (i, j) is pixel position, in which:
For image A (B image is similar), have:
A [i, j] represents the pixel value at (i, j),The average pixel value at (i, j) is represented,Represent (i,
J) direction the x average gradient value at place,The average gradient value in the direction y at (i, j) is represented,
Simultaneous following equations eliminate α1、β1、α2、β2, obtain the dx, dy compensated by deviation from linearity:
Two groups of equations can be by translating A image S in the direction x, y respectively afterwardsx、SyAfter a unit, solved with B image registration
Obtain (Sx、SyIt is taken as ± 1).
Experiment shows except in the case of image contains only direct current background, when image shift is less than 1 pixel dimension,
Arithmetic accuracy is up to 1/30 pixel dimension；For the image scene especially selected, precision is up to 1/50 pixel dimension or higher.Figure
As when being displaced to 2 pixel dimensions, Algorithm Error is very big.As a result bigger offset will have significantly away from phenomenon.
In above-mentioned steps 104, using Multi-resolution analyze in five plants of arrangement modes, by the image after image registration
B and image C synthesizes width combination image.As shown in figure 4, reflecting dislocation CCD obtains y (n after image1, n2) grid each point with
X (n before obtaining1, n2) grid each point corresponding relationship (4 points of a, b, c, d is only designated on figure).Viewpoint is analyzed from Multi-resolution
It sees, the variation of front and rear coordinate system is equivalent to coordinate system rotation, i.e., there is a dimension rotation matrix multiplier A between front and back:
In figure two kinds of distinct symbols △ andRespectively represent the low resolution of horizontal half of the pixel that all misplaces in vertical direction
Ccd image pixel center, grid intersection indicate combination image picture elements center.
Preferably, the low pass resolution filter H designed in the present invention0, the low-pass reconstruction filters G0, the height
Logical resolution filter H1And the high pass reconfigurable filter G1For the filter of the nondecomposable five plant shapes particular form of two dimension.
Low pass resolution filter H in the present invention0With low-pass reconstruction filters G0It designs by the following method:
One-dimensional filtering device is first expanded to two dimensional filter by separable method, then again by two by utilization level rotary process
It ties up filter and rotates 45 °, new filter is obtained, as shown in figure 5, wherein H0、H1、G0And G1Meet:
H0(z1,z2)H1(-z1,-z2)-H0(-z1,-z2)H1(z1,z2)=const
Using alternative transforms method, the construction two-dimensional linear phase filter since One Dimensional Polynomial matrix, the one of selection
Tie up polynomial matrix are as follows:
According to linear phase reconstruction filtering theorem, enable:
C=ad/a1
Obtain the shock response of filter are as follows:
Take a=1, a1=-4, a2=28, d=-4 are utilizedRelationship, obtain low pass
Resolution filter H0With low-pass reconstruction filters G0Matrix be respectively as follows:
Wherein,Respectively indicate 4 piecemeal squares of resolution filter
Battle array, a, d etc. are constants.
High pass resolution filter H in the present invention1With high pass reconfigurable filter G1It designs by the following method:
In conjunction with formula:
According to above-mentioned formula, in conjunction with the low pass resolution filter H0With the low-pass reconstruction filters G0Design method
It obtains.
In above-mentioned steps 105, interpolation processing is carried out to the combination image, by dislocation CCD two rows of on the combination image
Vacant grid point zero padding outside position, referring to Fig. 6.Symbol ◎ and zero respectively indicates former dislocation ccd data and zero-filled data in figure.Through
After overfill null transformation, the time domain and frequency domain of correspondence coordinate point data have variation:
The time domain of respective coordinates point data becomes:
The frequency domain of respective coordinates point data becomes:
Wherein, (n1,n2) representative image grid point, the position of x, y representation transformation front and back each grid point of image.To y (n1,
n2) carry out transform can obtain:
It enables in above formulaObtain the frequency domain relationship of respective coordinates point data are as follows:
In above-mentioned steps 106, to the combination image after the zero padding, the low pass resolution filter H of design is utilized0, high pass
Resolution filter H1, low-pass reconstruction filters G0With high pass reconfigurable filter G1It is decomposed and is rebuild, obtained high-resolution and try hard to
Picture, further are as follows:
Utilize the low pass resolution filter H of design0, high pass resolution filter H1, low-pass reconstruction filters G0It is reconstructed with high pass
Filter G1, in the low-pass reconstruction filters G0Matrix defined by within the scope of airspace, to the combination image after the zero padding
Low-pass filtering treatment and reconstruction are carried out, high resolution image is obtained.
Embodiment 3
A kind of Application Example of the invention presented below, referring to Fig. 7, the twin-line array imager based on Multi-resolution analysis
Part subpixel image synthetic method, comprising:
Step 201, using prismatic decomposition technology, be respectively staggered on CCD linear array direction and heading half of pixel respectively
Sampling obtains low resolution image A and low resolution image B, referring to fig. 2.
Wherein, specific acquisition process are as follows:, will be in the ccd detector according to the linear array direction and the heading
Heart coordinate is set to (0,0) and (0, p/2), and during push-scanning image, two CCD go to two width low resolution images after simultaneously
A and image B.
Step 202 carries out Histogram Matching to image A and image B, referring to Fig. 3.
Detailed process are as follows: first using image B as reference picture, histogram transformation is carried out to image A, so that the histogram of image A
Figure is consistent with the histogram of image B, obtains new images C, and in later interpolation processing, new images C is replaced the figure before transformation
As A.
Step 203, by be based on image First-order Gradient, make the sub- picture of variance and minimum between the image by linear interpolation
First registration Algorithm carries out image registration to image B and image C.
Step 204, image B and image C to the low resolution after registration utilize five be similar in Multi-resolution analysis
Strain arrangement (Quincunx) mode synthesizes width combination image.
Referring to fig. 4, which reflects dislocation CCD and obtains y (n after image1, n2) grid each point and x (n before obtaining1, n2) grid
The corresponding relationship (4 points of a, b, c, d is only designated on figure) of lattice each point.In terms of Multi-resolution analysis viewpoint, the variation of front and rear coordinate system
It is equivalent to coordinate system rotation, i.e., has a dimension rotation matrix multiplier A between front and back:
In figure two kinds of distinct symbols △ andRespectively represent the low resolution of horizontal half of the pixel that all misplaces in vertical direction
Ccd image pixel center, grid intersection indicate combination image picture elements center.
Step 205 carries out interpolation processing to the combination image, will be outside dislocation two rows of on the combination image position CCD
Vacant grid point zero padding illustrates the process referring to Fig. 6.
In Fig. 6, symbol ◎ and zero respectively indicates former dislocation ccd data and zero-filled data.After zero padding converts, front and back is right
The time domain of coordinate point data and frequency domain is answered to have variation:
The time domain of respective coordinates point data becomes:
The frequency domain of respective coordinates point data becomes:
Wherein, (n1,n2) representative image grid point, the position of x, y representation transformation front and back each grid point of image.To y (n1,
n2) carry out transform can obtain:
It enables in above formulaObtain the frequency domain relationship of respective coordinates point data are as follows:
Low pass, high pass resolution filter during step 206, design acquisition, zero padding: H0And H1, and in filtering reconstruction
Low pass, high pass reconfigurable filter in high resolution image process: G0And G1。
Step 207, the filter using design, in the low-pass reconstruction filters G0Matrix defined by airspace range
It is interior, low-pass filtering treatment and reconstruction are carried out to the combination image after the zero padding, obtain high resolution image.
As can be seen from the above embodiments beneficial effect existing for the application is:
First, the twin-line array image device subpixel image synthetic method of the present invention based on Multi-resolution analysis, from
The angle analysis integrally filtered the imaging problem of sub-pixed mapping, to entire image has all carried out effective analysis, theoretically can will point
It distinguishes that power rises to 1.42 times of original image, there is higher raising resolving power ability.
Second, the twin-line array image device subpixel image synthetic method of the present invention based on Multi-resolution analysis, benefit
, can also be with accuracy registration for noisy acoustic image with the information of entire image, therefore have the resolving power of noise image to raising
With better effect.
Third, the twin-line array image device subpixel image synthetic method of the present invention based on Multi-resolution analysis,
It is current to solve using the method for the over-sampling Data Synthesis high resolution image of acquisition in the case where not changing system parameter
Due to high resolution CCD it is deficient caused by the limited problem of picture resolution, greatly improve the effective utilization of system, reduce at
This.
It should be understood by those skilled in the art that, embodiments herein can provide as method, apparatus or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
Above description shows and describes several preferred embodiments of the present application, but as previously described, it should be understood that the application
Be not limited to forms disclosed herein, should not be regarded as an exclusion of other examples, and can be used for various other combinations,
Modification and environment, and the above teachings or related fields of technology or knowledge can be passed through within that scope of the inventive concept describe herein
It is modified.And changes and modifications made by those skilled in the art do not depart from spirit and scope, then it all should be in this Shen
It please be in the protection scope of appended claims.
Claims (7)
1. a kind of twin-line array image device subpixel image synthetic method based on Multi-resolution analysis characterized by comprising
In the imaging detector using line array CCD as representative, both threads battle array CCD is encapsulated in same focal plane device, or logical
The means for crossing optical spectroscopic are simultaneously imaged both threads battle array, and both threads battle array CCD is respectively staggered partly in linear array direction and heading
A pixel obtains two width low resolution image A and image B respectively；
Using described image B as reference picture, histogram transformation is carried out to described image A, so that the histogram of described image A and institute
The histogram for stating image B is consistent, the image C after being matched；
Image registration is carried out to described image B and described image C, further are as follows:
By being based on image First-order Gradient, make the sub-pixed mapping registration Algorithm pair of variance and minimum between the image by linear interpolation
Described image B and described image C carries out image registration, specifically:
Based on gradient G (Gx, Gy) formula are as follows:
Gx, GyGradient value on the direction x, y is respectively indicated, summation symbol is to i and j, and (i, j) is pixel position, in which:
For image A, have:
A [i, j] represents the pixel value at (i, j),The average pixel value at (i, j) is represented,It represents at (i, j)
The direction x average gradient value,The average gradient value in the direction y at (i, j) is represented,
Simultaneous following equations eliminate α1、β1、α2、β2, obtain the dx, dy compensated by deviation from linearity:
Two groups of equations can be by translating A image s in the direction x, y respectively afterwardsx、syAfter a unit, solve to obtain with B image registration,
sx、syIt is taken as ± 1；
Using Multi-resolution analyze in five plants of arrangement modes, by after image registration image B and image C synthesize a width
Combine image；
Interpolation processing is carried out to the combination image, vacant grid point outside dislocation two rows of on the combination image position CCD is mended
Zero；
Design low pass resolution filter H0, high pass resolution filter H1, low-pass reconstruction filters G0With high pass reconfigurable filter G1；
To the combination image after the zero padding, the low pass resolution filter H of design is utilized0, high pass resolution filter H1, low pass weight
Structure filter G0With high pass reconfigurable filter G1It is decomposed and is rebuild, obtain high resolution image.
2. the twin-line array image device subpixel image synthetic method according to claim 1 based on Multi-resolution analysis,
It is characterized in that,
Both threads battle array CCD is respectively staggered half of pixel in linear array direction and heading, obtains two width low resolution image A respectively
With image B, further are as follows:
Using the sub_pixel model means of prismatic decomposition technology, respectively it is staggered partly in the linear array direction of ccd detector and heading
A pixel is sampled, and according to the linear array direction and the heading, the ccd detector centre coordinate is set to
(0,0) and (0, p/2), by push-scanning image, both threads battle array CCD gets two width low resolution image A and image B simultaneously.
3. the twin-line array image device subpixel image synthetic method according to claim 1 based on Multi-resolution analysis,
It is characterized in that,
The low pass resolution filter H of design0, the low-pass reconstruction filters G0, the high pass resolution filter H1And institute
State high pass reconfigurable filter G1For the filter of the nondecomposable five plant shapes particular form of two dimension.
4. the twin-line array image device subpixel image synthetic method according to claim 1 based on Multi-resolution analysis,
It is characterized in that,
The low pass resolution filter H0With the low-pass reconstruction filters G0It designs by the following method:
One-dimensional filtering device is first expanded to two dimensional filter by separable method by utilization level rotary process, then again filters two dimension
Wave device rotates 45 °, obtains new filter, wherein H0、H1、G0And G1Meet:
H0(z1,z2)H1(-z1,-z2)-H0(-z1,-z2)H1(z1,z2)=const
Wherein, const is constant；Using alternative transforms method, the filter of two-dimensional linear phase is constructed since One Dimensional Polynomial matrix
Wave device, the One Dimensional Polynomial matrix of selection are as follows:
According to linear phase reconstruction filtering theorem, enable:
C=ad/a1
Obtain the shock response of filter are as follows:
Take a=1, a1=-4, a2=28, d=-4 are utilizedRelationship, obtain low pass decomposition
Filter H0With low-pass reconstruction filters G0Matrix be respectively as follows:
Wherein,Respectively indicate 4 matrixs in block form of resolution filter, a, d
It is constant.
5. the twin-line array image device subpixel image synthetic method according to claim 4 based on Multi-resolution analysis,
It is characterized in that,
The high pass resolution filter H1With the high pass reconfigurable filter G1It designs by the following method:
Utilize formula:
According to above-mentioned formula, in conjunction with the low pass resolution filter H0With the low-pass reconstruction filters G0Design method obtain.
6. the twin-line array image device subpixel image synthetic method according to claim 1 based on Multi-resolution analysis,
It is characterized in that,
Interpolation processing is carried out to the combination image, vacant grid point outside dislocation two rows of on the combination image position CCD is mended
After zero, the temporal relationship of respective coordinates point data are as follows:
Wherein, (n1,n2) representative image grid point, the position of x, y representation transformation front and back each grid point of image, to y (n1,n2) carry out
Transform can obtain:
It enables in above formulaObtain the frequency domain relationship of respective coordinates point data are as follows:
7. the twin-line array image device subpixel image synthetic method according to claim 1 based on Multi-resolution analysis,
It is characterized in that,
To the combination image after the zero padding, the low pass resolution filter H of design is utilized0, high pass resolution filter H1, low pass weight
Structure filter G0With high pass reconfigurable filter G1It is decomposed and is rebuild, obtain high resolution image, further are as follows:
Utilize the low pass resolution filter H of design0, high pass resolution filter H1, low-pass reconstruction filters G0With high pass reconstruction filtering
Device G1, in the low-pass reconstruction filters G0Matrix defined by within the scope of airspace, the combination image after the zero padding is carried out
Low-pass filtering treatment and reconstruction obtain high resolution image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610579045.8A CN106251288B (en) | 2016-07-21 | 2016-07-21 | Twin-line array image device subpixel image synthetic method based on Multi-resolution analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610579045.8A CN106251288B (en) | 2016-07-21 | 2016-07-21 | Twin-line array image device subpixel image synthetic method based on Multi-resolution analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106251288A CN106251288A (en) | 2016-12-21 |
| CN106251288B true CN106251288B (en) | 2019-04-09 |
Family
ID=57603233
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610579045.8A Active CN106251288B (en) | 2016-07-21 | 2016-07-21 | Twin-line array image device subpixel image synthetic method based on Multi-resolution analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106251288B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109870401B (en) * | 2019-02-28 | 2021-09-10 | 西安理工大学 | Flow cytometer and super-resolution cell image acquisition method |
| CN110211421B (en) * | 2019-06-27 | 2021-06-04 | 北京理工大学 | Cloud-based parking space information intelligent identification method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012156667A (en) * | 2011-01-25 | 2012-08-16 | Ricoh Co Ltd | Imaging apparatus and imaging method |
| CN104282007A (en) * | 2014-10-22 | 2015-01-14 | 长春理工大学 | Contourlet transformation-adaptive medical image fusion method based on non-sampling |
| CN105069748A (en) * | 2015-07-16 | 2015-11-18 | 哈尔滨工业大学 | Method for obtaining high-resolution image based on micro-satellite object scanning technique |
-
2016
- 2016-07-21 CN CN201610579045.8A patent/CN106251288B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012156667A (en) * | 2011-01-25 | 2012-08-16 | Ricoh Co Ltd | Imaging apparatus and imaging method |
| CN104282007A (en) * | 2014-10-22 | 2015-01-14 | 长春理工大学 | Contourlet transformation-adaptive medical image fusion method based on non-sampling |
| CN105069748A (en) * | 2015-07-16 | 2015-11-18 | 哈尔滨工业大学 | Method for obtaining high-resolution image based on micro-satellite object scanning technique |
Non-Patent Citations (3)
| Title |
|---|
| Contrast research on interpolation and subpixel imaging in CCD geometric superresolution reconstruction;Zheng-ping Xu;《2009 Asia Pacific Conference on Postgraduate Research in Microelectronics & Electronics (PrimeAsia)》;20090119;第294-297页 |
| 亚像元的CCD几何超分辨方法;徐正平;《光学精密工程》;20081215(第12期);第2447-2453页 |
| 基于流形学习和空间信息的改进N-FINDR端元提取算法;唐晓燕;《光谱学与光谱分析》;20130915(第9期);第2519-2524页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106251288A (en) | 2016-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tong et al. | Image registration with Fourier-based image correlation: A comprehensive review of developments and applications | |
| Ayhan et al. | A novel utilization of image registration techniques to process mastcam images in mars rover with applications to image fusion, pixel clustering, and anomaly detection | |
| US10353191B2 (en) | Circular scanning technique for large area inspection | |
| US7856154B2 (en) | System and method of super-resolution imaging from a sequence of translated and rotated low-resolution images | |
| US20110142366A1 (en) | System and Method for Super-Resolution Imaging from a Sequence of Color Filter Array (CFA) Low-Resolution Images | |
| WO2010122502A1 (en) | Super-resolution from a single signal | |
| Yurt et al. | Progressively volumetrized deep generative models for data-efficient contextual learning of MR image recovery | |
| CN102122359B (en) | Image registration method and device | |
| Sun et al. | Single image super-resolution using compressive sensing with a redundant dictionary | |
| CN106251288B (en) | Twin-line array image device subpixel image synthetic method based on Multi-resolution analysis | |
| Liu et al. | Multi-scale residual hierarchical dense networks for single image super-resolution | |
| CN104217412B (en) | Airborne super-resolution image reconstruction device and reconstruction method | |
| Zhou et al. | A general spatial-frequency learning framework for multimodal image fusion | |
| Shi et al. | Sub-pixel layout for super-resolution with images in the octic group | |
| Barnard et al. | High-resolution iris image reconstruction from low-resolution imagery | |
| CN102579045B (en) | Sparse imaging method for magnetic resonance over-sampling and under-sampling K data | |
| Mohammad-Djafari | Super-resolution: a short review, a new method based on hidden Markov modeling of HR image and future challenges | |
| Guo et al. | Super-resolution from unregistered aliased astronomical images | |
| Zhang et al. | Computational Super-Resolution Imaging With a Sparse Rotational Camera Array | |
| Barman et al. | GPU-accelerated adaptive dictionary learning and sparse representations for multispectral image super-resolution | |
| Lyakhov et al. | Single image Super-Resolution method based on bilinear interpolation and U-Net combination | |
| Seo et al. | A dual domain network for MRI reconstruction using gabor loss | |
| CN110503607A (en) | A kind of image split-joint method based on binocular vision | |
| CN113592716B (en) | Light field image space domain super-resolution method, system, terminal and storage medium | |
| Barman et al. | Edge-preserving single remote-sensing image super-resolution using sparse representations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |