CN104834088B - Wavefront coded imaging systems and the super-resolution processing method based on single image amplification - Google Patents
Wavefront coded imaging systems and the super-resolution processing method based on single image amplification Download PDFInfo
- Publication number
- CN104834088B CN104834088B CN201510166370.7A CN201510166370A CN104834088B CN 104834088 B CN104834088 B CN 104834088B CN 201510166370 A CN201510166370 A CN 201510166370A CN 104834088 B CN104834088 B CN 104834088B
- Authority
- CN
- China
- Prior art keywords
- eyeglass
- kng
- semiaperture
- super
- image
- 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
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B27/00—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00
- G02B27/0075—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00 with means for altering, e.g. increasing, the depth of field or depth of focus
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B27/00—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00
- G02B27/0012—Optical design, e.g. procedures, algorithms, optimisation routines
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B27/00—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00
- G02B27/0025—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00 for optical correction, e.g. distorsion, aberration
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B27/00—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00
- G02B27/50—Optics for phase object visualisation
- G02B27/52—Phase contrast optics
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B27/00—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00
- G02B27/58—Optics for apodization or superresolution; Optical synthetic aperture systems
Abstract
A kind of wavefront coded imaging systems and the super-resolution processing method based on single image amplification, including wavefront coded imaging lens, 1/3 inch of image detector and codec processing unit, wavefront coded imaging lens include the first eyeglass, phase mask plate, the second eyeglass and the 3rd eyeglass;First eyeglass, phase mask plate, the second eyeglass, the 3rd eyeglass, 1/3 inch of image detector and codec processing unit are successively set in same light path;First eyeglass, phase mask plate, the preceding surface of the second eyeglass and the 3rd eyeglass and the radius of curvature on rear surface, X-direction thang-kng semiaperture, Y-direction thang-kng semiaperture are different from each parameter of the prior art.The invention provides one kind on the premise of imaging sensor hardware condition is not changed, the blur-free imaging of super large depth of focus can be realized, can also obtain the wavefront coded imaging systems for the super-resolution reconstruction image for corresponding to smaller physics pixel dimension detector and the super-resolution processing method based on single image amplification.
Description
Technical field
The invention belongs to optical field, is related to a kind of wavefront coded imaging systems and the super-resolution based on single image amplification
Processing method, more particularly to it is a kind of using the Large depth-of-focus imaging system of wavefront coding technology and for wavefront coded imaging systems
Super-resolution image processing method.
Background technology
The depth of focus of expansion optical system always is the focus of academia's research, since the 1980s mid-term, though
Right panoramic method is proposed for depth of field extension, but until Univ Colorado-Boulder USA doctor Dowski and
Cathey is taught after the wavefront coded concept of nineteen ninety-five proposition, and field depth extending just has breakthrough truly.
By taking one dimensional optical system as an example, its defocus optical transfer function OTF can pass through the auto-correlation of generalized pupil function
Computing obtains, as follows:
Wherein, u and x is normalized spatial frequency and aperture plane lateral coordinates respectively;W20It is maximum defocus wave aberration
Coefficient;K is wave number;And f then represents phase-plate general expression.
For traditional imaging systems, the f items in above formula are not present, therefore can be readily available the specific of defocus OTF
Expression formula is:
It can be seen that when system is not introduced into phase-plate, its OTF is very sensitive to defocus, and can be empty in frequency
Between periodically there is zero point, so as to cause irreversible information loss.But once invented doctor E.R.Dowski
Cube phase-plate (f (x)=α x3) be incorporated on the entrance pupil face of optical system after, pass through static phase approximation method can
An entirely different defocus OTF is obtained, it is as follows:
It is clear that now defocus OTF mould, i.e. MTF are unrelated with defocus wave aberration coefficient, that is to say, that cube
Phase-plate can make system MTF insensitive to defocus;Although OTF phase bit position and degree of blur W20It is relevant, as long as but adjusting
Factor-alpha increase processed, it is to W20Dependency degree will significantly reduce.Simultaneously most importantly, with the addition of phase mask plate it
Afterwards, MTF simply has a certain degree of decline within effective frequency range in amplitude, and zero point or nearly zero point is not present, i.e.,
When defocus occurs in system, the information beyond primal system field depth is not lost, passes through digital image restoration algorithm afterwards
Can is effectively recovered.Simultaneously as phase-plate to the thang-kng amount and resolution ratio of system all without impacting, so ripple
Preceding coding is a kind of new big depth field imaging method for differing substantially from reduced bore method, central obscuration method or apodization.
Certainly, phase mask plate is the key for realizing that system optics transmission function OTF defocus is insensitive.However, such as
Fruit does not restore filtering, can not just obtain clearly big depth of focus image.Traditional recovery flow is only capable of obtaining and imaging sensor
The corresponding decoding image of valid pixel number, now the actual physical size of pixel determine what is characterized with every millimeter of demand pairs
System space resolution capability.It is most straight from the smaller detector of pixel if wanting the spatial resolution of further lifting system
The method connect.But the physics size of imaging sensor pixel is limited by factors such as manufacturing process and application conditions,
Can not possibly be ad infinitum small.Therefore, spatial resolution capability is realized on the premise of sensor physics pixel dimension is not changed
Lifting be one of focus of super-resolution research field.
The content of the invention
In order to solve above-mentioned technical problem present in background technology, the invention provides one kind not to change image sensing
On the premise of device hardware condition, it is not only able to realize the blur-free imaging of super large depth of focus, and can obtains and correspond to smaller physics
The wavefront coded imaging systems of the super-resolution reconstruction image of pixel dimension detector and the super-resolution based on single image amplification
Processing method.
The present invention technical solution be:The invention provides a kind of wavefront coded imaging systems, including it is wavefront coded
Imaging lens, 1/3 inch of image detector and codec processing unit, the wavefront coded imaging lens include the first eyeglass,
Phase mask plate, the second eyeglass and the 3rd eyeglass;First eyeglass, phase mask plate, the second eyeglass, the 3rd eyeglass, 1/3
Inch image detector and codec processing unit are successively set in same light path;It is characterized in that:
The radius of curvature on the preceding surface of first eyeglass is 18.91mm, and the X-direction thang-kng on the preceding surface of the first eyeglass is partly
Aperture and Y-direction thang-kng semiaperture are 6.22mm;The radius of curvature on the rear surface of first eyeglass is 234mm, first
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the rear surface of eyeglass are 5.85mm;The preceding table of first eyeglass
The distance between face and the rear surface of the first eyeglass, i.e. the center thickness of the first eyeglass is 3.0mm;
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the preceding surface of the phase mask plate are 4.95mm;
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the rear surface of the phase mask plate are 4.42mm;Described first
The distance between the distance between the rear surface of eyeglass and the preceding surface of phase mask plate, i.e. the first eyeglass and phase mask plate are
3.22mm;The distance between the preceding surface of the phase mask plate and the rear surface of phase mask plate, the i.e. thickness of phase mask plate
Degree is 2.0mm;
The radius of curvature on the preceding surface of second eyeglass is -34.973mm, the X-direction thang-kng on the preceding surface of the second eyeglass
Semiaperture and Y-direction thang-kng semiaperture are 4.43mm;The radius of curvature on the rear surface of second eyeglass is 15.776mm,
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the rear surface of the second eyeglass are 5.46mm;The phase mask plate
The distance between the preceding surface of rear surface and the second eyeglass, i.e., the distance between phase mask plate and the second eyeglass are 2.0mm;
The distance between the preceding surface of second eyeglass and the rear surface of the second eyeglass, i.e. the center thickness of the second eyeglass is 3.0mm;
The radius of curvature on the preceding surface of the 3rd eyeglass is 33.6240mm, the X-direction thang-kng on the preceding surface of the 3rd eyeglass
Semiaperture and Y-direction thang-kng semiaperture are 5.57mm;The radius of curvature on the rear surface of the 3rd eyeglass is -26.53mm,
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the rear surface of the 3rd eyeglass are 4.24mm;Second eyeglass
The distance between surface and the preceding surface of the 3rd eyeglass afterwards, i.e., the distance between the second eyeglass and the 3rd eyeglass are 5.17mm;Institute
The distance between the preceding surface of the 3rd eyeglass and the rear surface of the 3rd eyeglass are stated, i.e. the center thickness of the 3rd eyeglass is 3.0mm;
The distance between the rear surface of 3rd eyeglass and 1/3 inch of image detector are 39.3135mm.
The 2D mask functional forms of above-mentioned phase mask plate are:
Wherein:
α characterizes the phase-modulation intensity of square phase mask plate three times, and the α values are 0.0130mm;
X and y is normalized aperture coordinate, unit mm, x the and y spans be [- 4.9513,
4.9513]。
The focal length of above-mentioned wavefront coded imaging lens is 50mm, relative aperture 1:4.5, full filed angle is approximately 10 °, work spectrum
Section 480um~680um.
A kind of super-resolution processing method based on single image amplification based on wavefront coded imaging systems as described above,
It is characterized in that:The super-resolution processing method based on single image amplification comprises the following steps:
1) the size p of the actual pixel of sensor obtained desired by super-resolution is selectednew, and determine therefrom that super-resolution net
Table images Y and decoding are exactly before in the proportionality coefficient between uniformly fuzzy original-resolution image I, i.e., described proportionality coefficient
The multiplication factor of target image, the multiplication factor of the target image is δ=[p/pnew],
Wherein:
[] represents and rounded;
P then represent with before decoding in the corresponding sensor pixels sizes of uniformly fuzzy original-resolution image I;
2) line direction and column direction before decoding in uniformly fuzzy original-resolution image I are amplified δ times, this
When super-resolution grid image Y valid pixel number will be changed into before decoding in uniformly fuzzy original-resolution image I δ2
Times;
3) there is smaller pixel figure using what is obtained desired by the gain of parameter of wavefront coded imaging lens as implied above
As the original point spread function Psf before sensor sampleoptical;
4) according to the multiplication factor δ of target image, by reversely filling resampling, Psf is utilizedopticalCalculate and correspond to more
Small sensor actual physics pixel pnewSampling point spread function Psfcalculated;
5) with PsfcalculatedAs priori, target image is amplified to line direction and column direction and put
The blurred picture of big δ times of multiple carries out recovery filtering, realizes the super-resolution rebuilding based on single image amplification.
Above-mentioned steps 2) in will decode before it is equal in uniformly fuzzy original-resolution image I line direction and column direction
Used interpolation method is closest interpolation when amplifying δ times.
Above-mentioned steps 3) in obtain desired by obtain have smaller pixel imaging sensor sampling before original point expand
Dissipate function PsfopticalMode be to be exported by optical system design software.
Above-mentioned multiplication factor is not more than 4.
It is an advantage of the invention that:
The present invention is on the basis of wavefront coded imaging systems physical characteristic is furtherd investigate, it is proposed that one kind is suitably applied
Wavefront Coding system, based on image amplification super-resolution processing method, on the premise of detector pixel dimension is not changed,
It can not only realize that super large depth of focus is imaged, and the purpose of super-resolution imaging can be reached.Image amplification is simplest incites somebody to action
The method that low-resolution image is mapped to fine-resolution meshes.Traditional amplification of the image based on interpolation can not bring information content
Increase, and simply merely add number of pixels, if dealt with improperly, the resolved bands of details can also be come with unfavorable shadow
Ring.Super-resolution processing method proposed by the invention is not only able to increase image using wavefront coded imaging mechanism as theoretical foundation
Valid pixel number, and traditional images can be eliminated and amplify destruction to local detail, such as edge feature, be to wavefront
The deep excavation of the potential characteristic of coded imaging technology.Accordingly, the present invention devises the wavefront for carrying classical phase-plate square three times
Coded imaging systems are used for verification experimental verification.As a result show, this super-resolution processing method based on wavefront coded mechanism is in office
There is great advantage in terms of the accurate description of portion's amplification details.
Brief description of the drawings
Fig. 1 is the principle schematic of wavefront coded Large depth-of-focus imaging system;
Fig. 2 is the optical delivery that wavefront coded Large depth-of-focus imaging system corresponds to different image-forming ranges (under different defocusing amounts)
Function;
Fig. 3 is that optical point spread function is corresponded to not by what analog sampling was calculated as derived from optical design software
With the sampling point spread function of sensor physics Pixel size;
Fig. 4 be wavefront coded Large depth-of-focus imaging system material object and under parallel light tube through 5.2um Pixel size figures
The actual point spread function obtained as sensor sample;
Fig. 5 is the super-resolution processing flow chart provided by the present invention for wavefront coded imaging systems;
Fig. 6 be wavefront coded Large depth-of-focus imaging system one group of field trial result and corresponding super-resolution processing after
Partial result.
Embodiment
The present invention relates to Large depth-of-focus imaging system of the complete set based on wavefront coded mechanism and for wavefront coded
The super-resolution processing method based on single image amplification of imaging system.
The invention provides a kind of wavefront coded imaging systems, including wavefront coded imaging lens, 1/3 inch of image detection
Device and codec processing unit, wavefront coded imaging lens include the first eyeglass (eyeglass 1 in Fig. 1), phase mask plate, second
Eyeglass (eyeglass 2 in Fig. 1) and the 3rd eyeglass (eyeglass 3 in Fig. 1);First eyeglass, phase mask plate, the second eyeglass,
Three eyeglasses, 1/3 inch of image detector and codec processing unit are successively set in same light path;
The radius of curvature on the preceding surface of the first eyeglass is 18.91mm, the X-direction thang-kng semiaperture on the preceding surface of the first eyeglass
And Y-direction thang-kng semiaperture is 6.22mm;The radius of curvature on the rear surface of the first eyeglass is 234mm, after the first eyeglass
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on surface are 5.85mm;The preceding surface of first eyeglass and the first eyeglass
The distance between rear surface, i.e. the center thickness of the first eyeglass is 3.0mm;
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the preceding surface of phase mask plate are 4.95mm;Phase
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the rear surface of mask plate are 4.42mm;The rear surface of first eyeglass
The distance between preceding surface of phase mask plate, i.e. the distance between the first eyeglass and phase mask plate are 3.22mm;Phase
The distance between the preceding surface of mask plate and the rear surface of phase mask plate, i.e. the thickness of phase mask plate is 2.0mm;
The radius of curvature on the preceding surface of the second eyeglass is -34.973mm, the X-direction thang-kng half bore on the preceding surface of the second eyeglass
Footpath and Y-direction thang-kng semiaperture are 4.43mm;The radius of curvature on the rear surface of the second eyeglass is 15.776mm, the second eyeglass
Rear surface X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture be 5.46mm;The rear surface of phase mask plate and the
The distance between preceding surface of two eyeglasses, i.e. the distance between phase mask plate and the second eyeglass are 2.0mm;Before second eyeglass
The distance between surface and the rear surface of the second eyeglass, i.e. the center thickness of the second eyeglass is 3.0mm;
The radius of curvature on the preceding surface of the 3rd eyeglass is 33.6240mm, the X-direction thang-kng half bore on the preceding surface of the 3rd eyeglass
Footpath and Y-direction thang-kng semiaperture are 5.57mm;The radius of curvature on the rear surface of the 3rd eyeglass is -26.53mm, the 3rd eyeglass
Rear surface X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture be 4.24mm;The rear surface and the 3rd of second eyeglass
The distance between preceding surface of eyeglass, i.e., the distance between the second eyeglass and the 3rd eyeglass are 5.17mm;The preceding table of 3rd eyeglass
The distance between face and the rear surface of the 3rd eyeglass, i.e. the center thickness of the 3rd eyeglass is 3.0mm;
The distance between the rear surface of 3rd eyeglass and 1/3 inch of image detector are 39.3135mm.
The 2D mask functional forms of phase mask plate are:
Wherein:
α characterizes the phase-modulation intensity of square phase mask plate three times, and α values are 0.0130mm;
X and y is normalized aperture coordinate, and unit mm, x and y span is [- 4.9513,4.9513].
The focal length of wavefront coded imaging lens is 50mm, relative aperture 1:4.5, full filed angle is approximately 10 °, operating spectrum band
480um~680um.
The present invention additionally provides one kind and is based on the wavefront coded imaging while as above wavefront coded imaging systems are provided
System based on single image amplification super-resolution processing method, based on single image amplification super-resolution processing method include with
Lower step:
1) the size p of the actual pixel of sensor obtained desired by super-resolution is selectednew, and determine therefrom that super-resolution net
Before in the proportionality coefficient between uniformly fuzzy original-resolution image I, i.e. proportionality coefficient is exactly target for table images Y and decoding
The multiplication factor of image, the multiplication factor of target image is δ=[p/pnew],
Wherein:
[] represents and rounded;
P then represent with before decoding in the corresponding sensor pixels sizes of uniformly fuzzy original-resolution image I;
2) line direction and column direction before decoding in uniformly fuzzy original-resolution image I are amplified δ times, this
When super-resolution grid image Y valid pixel number will be changed into before decoding in uniformly fuzzy original-resolution image I δ2
Times;The interpolation method of use is closest interpolation;
3) there is smaller pixel figure using what is obtained desired by the gain of parameter of wavefront coded imaging lens as implied above
As the original point spread function Psf before sensor sampleoptical;What is obtained desired by being obtained in step 3) has smaller pixel
Original point spread function Psf before imaging sensor samplingopticalMode be to be exported by optical system design software.
4) according to the multiplication factor δ of target image, by reversely filling resampling, Psf is utilizedopticalCalculate and correspond to more
Small sensor actual physics pixel pnewSampling point spread function Psfcalculated;
5) with PsfcalculatedAs priori, target image is amplified to line direction and column direction and put
The blurred picture of big δ times of multiple carries out recovery filtering, realizes the super-resolution rebuilding based on single image amplification.
With reference to figure 1 (the prototype framework of wavefront coded Large depth-of-focus imaging system), in system proposed by the invention,
After imageable target 1 is by conventional wavefront coded imaging lens 2, fuzzy intermediary image is formed on 1/3 inch of image detector 3,
Then codec processing unit 4 carries out deconvolution processing, finally gives the big depth of focus image of focus.It is original except that can obtain
Outside the big depth of focus image of resolution ratio, codec processing unit 4 can also implement the super-resolution based on single image amplification as needed
Rate is rebuild, and the image of smaller physics pixel dimension sensor is corresponded to by being calculated.
Wavefront coded imaging lens proposed by the invention are to carry to have three on the basis of classical Cook three lens objectives
What the phase mask plate of secondary square phase distribution obtained.Three conventional spherical lenses are using in general glass lens processing technology
It can obtain, square phase mask plate then needs to be completed by five degree of freedom Free-Form Surface Machining equipment three times.Target scene is sent
Imaging light after wavefront coded imaging lens, it has the depth information of the scene diverse location of carrying and compiled
Code, the thing followed be system to the insensitive of defocus, by Fig. 2, (wavefront coded Large depth-of-focus imaging system corresponds to different imagings
Shown in distance (5m (a), 7m (b), 10m (c), 50m (d) and infinity (e)).Wherein abscissa is represented with every millimeter of demand pairs
The spatial frequency of sign, ordinate then represent the MTF amplitudes after normalization.) shown in modulation transfer function reflect.
It can be seen that the MTF corresponding to different defocusing amounts, different visual fields and different wave length is respectively provided with fabulous uniformity.Although compile
Mtf value after code is less than the value before coding, but it is not in null value in the case where defocusing amount be present, therefore will not make
Into the loss of image detail information.It is that the fuzzy image of uniformity is presented in a width on image detector, uses codec processing unit
It is decoded, the mtf value of system is promoted to close to diffraction limited, so as to recover sharp keen clearly image.
Wavefront coded imaging systems involved in the present invention are before coming into operation, it is necessary to enter with the help of parallel light tube
Row staking-out work.Research shows:For the phase mask plate that rectangle can divide, point spread function corresponding to it is simultaneously non-rotating
Symmetrically, its main energetic is distributed in orthogonal x directions and y directions.If the x-axis y of wavefront coded imaging lens point spread function
Axle can not be well corresponding with the x-axis y-axis of imaging sensor, then strong fake information will be produced in restored image.Therefore,
Phase mask plate designed by the present invention is arranged on a gripper frame with mechanical rotation apparatus (as shown in Figure 4), can be with
Rotation of the phase mask plate centered on optical axis is realized in the range of 0~90 °.This structure can just accomplish to make phase-plate
Two axles accurately align with the axle of imaging sensor two.
In view of the particularity of wavefront coding technology, the present invention proposes a kind of Super-resolution reconstruction based on single image amplification
Processing method is built, the reconstruction figure corresponding to smaller physics pixel can be obtained on the premise of imaging sensor hardware is not changed
Picture.The it is proposed of this method is based on following results of study.
Any one imaging lens is after by good aberration correction, the branch of its optical point spread function
Holding domain will closely.Now, when simulating the system point spread function after detector samples using optical point spread function, no
With the difference very little between the result corresponding to Pixel size detector.However, because the optical point of Wavefront Coding system spreads
Function has great support region, so the difference of Pixel size will produce completely different system point spread function.Pixel is got over
Small, sampling point spread function is just closer to optical point spread function.In other words, pixel is smaller, and sampling point spread function is to being
The description of system is more accurate.As Fig. 3 (Fig. 3 give it is corresponding with the wavefront coded imaging lens, exported by optical design software
Optical point spread function by analog sampling be calculated correspond to different sensors physics Pixel size (5.2um,
3.2um, 2.6um and 1.7um) sampling point spread function) shown in, by simulate calculate, led by optical system design software
The optical point spread function gone out can obtain and different size pixel sensor (5.2um, 3.2um, 2.6um and 1.7um) phase
Corresponding sampling point spread function.By comparing Fig. 3, (Fig. 4 gives the material object of the wavefront coded Large depth-of-focus imaging system with Fig. 4
And sample obtained actual point spread function through 5.2um Pixel size imaging sensors under parallel light tube), it is not difficult to send out
Existing, point spread function corresponding to 5.2um pixels caused by simulation has very high with the actual measurement point spread function under parallel light tube
Similarity.Therefore, after using optical system design software extraction optical point spread function, it is possible to be more precisely calculated
Actual point spread function after the sampling of different size pixel, the result of study just obtain not change sensor hardware condition
Theoretical foundation has been established corresponding to the actual samples point spread function of smaller pixel.
This characteristic based on wavefront coded imaging systems, the super-resolution proposed by the present invention for Wavefront Coding system
Handling process is as shown in Figure 5.First, it is determined that desired obtained detector physics Pixel size, and by with current sensor
The comparison of actual Pixel size obtains the enlargement ratio of target image.For example the Pixel size of present image sensor is 5.2um,
Assuming that go for the imaging effect of 2.6um sensors, then now the enlargement ratio of target image is equal to 2.Secondly, by light
Learn point spread function to set out, obtaining actual samples point corresponding with the detector with required Pixel size by analog sampling expands
Dissipate function.Again, the fuzzy intermediate image after being encoded through phase-plate is amplified to corresponding multiple.Finally, counted using simulation
The actual samples point spread function corresponding with smaller pixel detector obtained is calculated, passes through the convolution combination Richardson- that does
Lucy algorithms carry out recovery reconstruction to the middle blurred picture after amplification, so as to obtain the effect of Super-resolution Reconstruction.
Fig. 6 gives one group of field trial knot carried out using wavefront coded super-resolution imaging system proposed by the present invention
Fruit, wherein, the sensor physics Pixel size corresponding to original-resolution image is 5.2um.It can be seen that the figure after decoding
As very clear, and the edge in scene all significantly enhances, because phase mask plate is eliminating defocus sensitiveness
While wavefront distortion also to system serve inhibitory action, so the filtering decoding image of Wavefront Coding system is due to eliminating
Blur degradation that wavefront distortion is brought and generate the effect of scene enhancing.Using the super-resolution rebuilding flow described in Fig. 5,
Fig. 6 also gives corresponding processing example, wherein it is 1.3um's that the treatment effect of 4 times of super-resolutions, which is equivalent to by Pixel size,
The image that imaging sensor is captured.It can be seen that compared with simple amplification, super-resolution processing method proposed by the invention
The effect after amplification can be significantly increased, eliminates the sawtooth effect for routinely amplifying and bringing, can be more accurately to scene information
It is described.This is turned out, and the superresolution processing flow for Wavefront Coding system proposed by the invention is very effective.
However, it is contemplated that the problems such as energy deficiency and signal to noise ratio that small pixel is brought reduce, the superresolution processing flow institute shown in Fig. 5
The multiplication factor that can be obtained is in the presence of limitation.Research shows, what super-resolution processing flow proposed by the invention was applicable
Multiplication factor is usually no more than 4.
Claims (5)
1. a kind of wavefront coded imaging systems, including wavefront coded imaging lens, 1/3 inch of image detector and decoding process
Unit, the wavefront coded imaging lens include the first eyeglass, phase mask plate, the second eyeglass and the 3rd eyeglass;Described
One eyeglass, phase mask plate, the second eyeglass, the 3rd eyeglass, 1/3 inch of image detector and codec processing unit are set gradually
In same light path;It is characterized in that:
The radius of curvature on the preceding surface of first eyeglass is 18.91mm, the X-direction thang-kng semiaperture on the preceding surface of the first eyeglass
And Y-direction thang-kng semiaperture is 6.22mm;The radius of curvature on the rear surface of first eyeglass is 234mm, the first eyeglass
Rear surface X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture be 5.85mm;The preceding surface of first eyeglass with
The distance between rear surface of first eyeglass is 3.0mm;
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the preceding surface of the phase mask plate are 4.95mm;It is described
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the rear surface of phase mask plate are 4.42mm;First eyeglass
The distance between the preceding surface of rear surface and phase mask plate be 3.22mm;Covered with phase on the preceding surface of the phase mask plate
The distance between rear surface of diaphragm plate is 2.0mm;
The radius of curvature on the preceding surface of second eyeglass is -34.973mm, the X-direction thang-kng half bore on the preceding surface of the second eyeglass
Footpath and Y-direction thang-kng semiaperture are 4.43mm;The radius of curvature on the rear surface of second eyeglass is 15.776mm, second
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the rear surface of eyeglass are 5.46mm;After the phase mask plate
The distance between surface and the preceding surface of the second eyeglass are 2.0mm;The preceding surface of second eyeglass and the rear table of the second eyeglass
The distance between face is 3.0mm;
The radius of curvature on the preceding surface of the 3rd eyeglass is 33.6240mm, the X-direction thang-kng half bore on the preceding surface of the 3rd eyeglass
Footpath and Y-direction thang-kng semiaperture are 5.57mm;The radius of curvature on the rear surface of the 3rd eyeglass is -26.53mm, the 3rd
The X-direction thang-kng semiaperture and Y-direction thang-kng semiaperture on the rear surface of eyeglass are 4.24mm;The rear table of second eyeglass
The distance between face and the preceding surface of the 3rd eyeglass are 5.17mm;The preceding surface of 3rd eyeglass and the rear surface of the 3rd eyeglass
The distance between be 3.0mm;
The distance between the rear surface of 3rd eyeglass and 1/3 inch of image detector are 39.3135mm;
The 2D mask functional forms of the phase mask plate are:
<mrow>
<mi>Q</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>&alpha;</mi>
<mo>&CenterDot;</mo>
<mrow>
<mo>(</mo>
<msup>
<mrow>
<mo>(</mo>
<mfrac>
<mi>x</mi>
<mn>4.9513</mn>
</mfrac>
<mo>)</mo>
</mrow>
<mn>3</mn>
</msup>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<mfrac>
<mi>y</mi>
<mn>4.9513</mn>
</mfrac>
<mo>)</mo>
</mrow>
<mn>3</mn>
</msup>
<mo>)</mo>
</mrow>
</mrow>
Wherein:
α characterizes the phase-modulation intensity of square phase mask plate three times, and the α values are 0.0130mm;
X and y is normalized aperture coordinate, and unit mm, x the and y spans are [- 4.9513,4.9513].
2. wavefront coded imaging systems according to claim 1, it is characterised in that:Jiao of the wavefront coded imaging lens
Away from being 50mm, relative aperture 1:4.5, full filed angle is approximately 10 °, operating spectrum band 480um~680um.
3. a kind of wavefront coded imaging systems based on as described in claim 1-2 any claims are put based on single image
Big super-resolution processing method, it is characterised in that:The super-resolution processing method based on single image amplification includes following step
Suddenly:
1) the actual Pixel size p of sensor obtained desired by super-resolution is selectednew, and determine therefrom that super-resolution grid image Y
With being in the proportionality coefficient between uniformly fuzzy original-resolution image I before decoding, i.e., described proportionality coefficient is exactly target figure
The multiplication factor of picture, the multiplication factor of the target image is δ=[p/pnew],
Wherein:
[] represents and rounded;
P then represent with before decoding in the corresponding sensor pixels sizes of uniformly fuzzy original-resolution image I;
2) line direction and column direction before decoding in uniformly fuzzy original-resolution image I are used into closest interpolation
Method amplifies δ times, and now super-resolution grid image Y valid pixel number will be original in what is uniformly obscured before being changed into decoding
Image in different resolution I δ2Times;
3) using obtaining desired by the gain of parameter of the wavefront coded imaging lens described in claim 1-2 any claims
Original point spread function Psf before being sampled with smaller pixel imaging sensoroptical;
4) according to the multiplication factor δ of target image, by reversely filling resampling, Psf is utilizedopticalCalculate corresponding to smaller
The actual Pixel size p of sensornewSampling point spread function Psfcalculated;
5) with PsfcalculatedAs priori, the times magnification of target image has been amplified to line direction and column direction
The blurred picture of δ times of number carries out recovery filtering, realizes the super-resolution rebuilding based on single image amplification.
4. the super-resolution processing method according to claim 3 based on single image amplification, it is characterised in that:The step
3) what is obtained desired by being obtained in has the original point spread function Psf before the sampling of smaller pixel imaging sensoroptical's
Mode is exported by optical system design software.
5. the super-resolution processing method based on single image amplification according to claim 3 or 4, it is characterised in that:It is described
Multiplication factor is not more than 4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510166370.7A CN104834088B (en) | 2015-04-09 | 2015-04-09 | Wavefront coded imaging systems and the super-resolution processing method based on single image amplification |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510166370.7A CN104834088B (en) | 2015-04-09 | 2015-04-09 | Wavefront coded imaging systems and the super-resolution processing method based on single image amplification |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104834088A CN104834088A (en) | 2015-08-12 |
CN104834088B true CN104834088B (en) | 2017-12-05 |
Family
ID=53812077
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510166370.7A Active CN104834088B (en) | 2015-04-09 | 2015-04-09 | Wavefront coded imaging systems and the super-resolution processing method based on single image amplification |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104834088B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018218063A1 (en) | 2017-05-24 | 2018-11-29 | The Trustees Of Columbia University In The City Of New York | Broadband achromatic flat optical components by dispersion-engineered dielectric metasurfaces |
CA3074566A1 (en) | 2017-08-31 | 2019-03-07 | Metalenz, Inc. | Transmissive metasurface lens integration |
CN116171399A (en) * | 2020-07-31 | 2023-05-26 | 华为技术有限公司 | Optical device |
CN113031260A (en) * | 2021-03-12 | 2021-06-25 | 中国人民解放军国防科技大学 | Wavefront coding imaging system based on arcsine type phase mask plate |
CN113393380B (en) * | 2021-05-27 | 2024-04-09 | 中国科学院西安光学精密机械研究所 | High-resolution imaging camera with large F number diffraction real-time correction based on pixel coding |
US11927769B2 (en) | 2022-03-31 | 2024-03-12 | Metalenz, Inc. | Polarization sorting metasurface microlens array device |
CN116681595B (en) * | 2023-08-01 | 2023-11-03 | 长春理工大学 | Remote computing super-resolution imaging device based on multimodal PSF |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003009041A2 (en) * | 2001-07-20 | 2003-01-30 | The Regents Of The University Of Colorado | Wavefront coding zoom lens imaging systems |
CN101449193A (en) * | 2006-03-06 | 2009-06-03 | 全视Cdm光学有限公司 | Zoom lens systems with wavefront coding |
CN102243329A (en) * | 2010-05-14 | 2011-11-16 | 中国科学院西安光学精密机械研究所 | Phase plate for wave-front coding imaging technology |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN204719330U (en) * | 2015-04-09 | 2015-10-21 | 中国科学院西安光学精密机械研究所 | Wavefront coded imaging systems |
-
2015
- 2015-04-09 CN CN201510166370.7A patent/CN104834088B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003009041A2 (en) * | 2001-07-20 | 2003-01-30 | The Regents Of The University Of Colorado | Wavefront coding zoom lens imaging systems |
CN101449193A (en) * | 2006-03-06 | 2009-06-03 | 全视Cdm光学有限公司 | Zoom lens systems with wavefront coding |
CN102243329A (en) * | 2010-05-14 | 2011-11-16 | 中国科学院西安光学精密机械研究所 | Phase plate for wave-front coding imaging technology |
Also Published As
Publication number | Publication date |
---|---|
CN104834088A (en) | 2015-08-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104834088B (en) | Wavefront coded imaging systems and the super-resolution processing method based on single image amplification | |
CN104834089B (en) | Wavefront coded imaging systems and super-resolution processing method | |
CN204719330U (en) | Wavefront coded imaging systems | |
US9251565B2 (en) | Hyper-resolution imaging | |
US8471905B2 (en) | Methods and apparatus for imaging | |
US20120201477A1 (en) | Kinetic Super-Resolution Imaging | |
AU2005283143A1 (en) | Multifocal lens with extended depth of field | |
CN110544205A (en) | Image super-resolution reconstruction method based on visible light and infrared cross input | |
Ng et al. | Digital correction of lens aberrations in light field photography | |
JP2013531268A (en) | Measuring distance using coded aperture | |
CN102170526A (en) | Method for calculation of defocus fuzzy core and sharp processing of defocus fuzzy image of defocus fuzzy core | |
WO2023138629A1 (en) | Encrypted image information obtaining device and method | |
Sanghvi et al. | Photon limited non-blind deblurring using algorithm unrolling | |
Vint et al. | Analysis of deep learning architectures for turbulence mitigation in long-range imagery | |
JP2015004883A (en) | Image forming optical system, imaging system, and imaging method | |
CN204613515U (en) | A kind of wavefront coded imaging systems | |
Shima et al. | Comparison of image alignment on hexagonal and square lattices | |
CN105243677B (en) | A kind of real-time unimolecule localization method and system guaranteeing precision | |
CN114004770B (en) | Method and device for accurately correcting satellite space-time diagram and storage medium | |
CN116704590A (en) | Iris image correction model training method, iris image correction device and iris image correction medium | |
Luo et al. | Midfrequency-based real-time blind image restoration via independent component analysis and genetic algorithms | |
Maik et al. | Blind deconvolution using maximum a posteriori (MAP) estimation with directional edge based priori | |
Inzerillo | Super-resolution images on mobile smartphone aimed at 3D modeling | |
Flusser et al. | Invariants to convolution in arbitrary dimensions | |
Tang et al. | Motion deblurring based on local temporal compressive sensing for remote sensing image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
EXSB | Decision made by sipo to initiate substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |