CN109828444A - A kind of auto focusing method for the measurement of multiple target objects digital hologram - Google Patents
A kind of auto focusing method for the measurement of multiple target objects digital hologram Download PDFInfo
- Publication number
- CN109828444A CN109828444A CN201910244991.0A CN201910244991A CN109828444A CN 109828444 A CN109828444 A CN 109828444A CN 201910244991 A CN201910244991 A CN 201910244991A CN 109828444 A CN109828444 A CN 109828444A
- Authority
- CN
- China
- Prior art keywords
- hologram
- measurement
- holographic
- target objects
- multiple target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Holo Graphy (AREA)
Abstract
A kind of auto focusing method for the measurement of multiple target objects digital hologram, it is related to digital hologram field of measuring technique, the focusing curve peak value that criterion function can be made to obtain due to the influence of system and ambient noise in the prior art for solution is lower, it is not easy the problem of differentiating, the present invention includes step 1: acquisition hologram；Step 2: separation hologram；Step 3: setting automatic seeking coke search range, reconstruction image；Step 4: automatic focusing curve is acquired；Step 5: using the automatic seeking coke search range of setting as abscissa, using Ci value as ordinate, the Ci Drawing of Curve that all holographic subgraphs are obtained is under unified coordinate system；Step 6: the specific reproduction distance of the multiple objects obtained according to step 5 rebuilds original hologram, and all reconstructions are obtained the spatial distribution state of multiple target objects as being placed in uniform spaces coordinate system.The present invention makes the peak value at focusing more obvious automatic focusing criterion function by exponential transform, and the effective peak value that reduces distinguishes difficulty.
Description
Technical field
It is specially a kind of for the poly- of multiple target objects digital hologram measurement the present invention relates to digital hologram field of measuring technique
Burnt method.
Background technique
The amplitude and phase information of testee can be obtained in digital hologram measurement, is the important measurement hand in optical measurement
Section.Using the hologram of electrooptical device (CCD or CMOS) record testee, by hologram by that can be obtained after digital processing
Obtain the reproduction distance and reproduction image of measured object.Therefore digital hologram can realize the real-time measurement and positioning of target surface pattern.
The optical field distribution of reproduction image is calculated using the diffraction integral of light propagation for the restructuring procedure of hologram reconstruction,
It is the key that rebuild reproduction image quality that distance is accurately rebuild in middle acquisition.Therefore, it is accurately reproduced apart from state to obtain
Inside and outside scientific research personnel proposes a variety of methods focused automatically, mainly includes gradient of image and gray scale algorithm, the calculation of image grayscale variance
The methods of method, comentropy and high frequency criterion.
Patent document CN106502074A in 2017 discloses a kind of automatic focusing side of image planes digital holographic micro-measuring
Method.This method obtains the hologram before and after focal position before measuring and the auto-correlation operator of every width figure is obtained by calculation, and leads to
Over-fitting finds out the maximum value of auto-correlation operator, the i.e. corresponding focusing surface of the value.This method provides new real-time measurement sample tables
Great convenience that the method for face form has been post-processing band, but control the process phase that motorized subject table shoots multiple holograms
To cumbersome so that error increases.
These algorithms and method have preferable processing result for the automatic focusing of single target object under certain condition, but
For the treatment effects focused automatically of multiple objects, there is no more reports.It is difficult to ensure that measurement in actual measurement environment
There was only simple target object in the process, it is therefore desirable to more to study and inquire into the method that multiple target objects focus automatically to solve reality
Needs in measurement.
2010 Nian Xuyuan it is strong et al. it is entitled " particle segmentation identification in digital in-line holographic particle field detecting with it is independent poly-
Intensity fluctuation mean square deviation is recycled to realize as criterion function by splitting the object in hologram in the paper of coke "
Focusing to multiple objects.But processing result finds that the peak value of focusing curve is not obvious, if seek burnt range it is larger when
It is likely to so that characteristic peak is submerged in noise.
To sum up, it can be found that when finding object in digital hologram using criterion function, unimodal song is presented in single object
Line, the abscissa at peak value are focal position, since the influence of system and ambient noise can be such that criterion function obtains in experiment
Focusing curve peak value it is lower, be not easy to differentiate.Meanwhile when the hologram of record is there are when multiple objects, each target position
Locate homologous thread on peak value it is lower, in addition noise present in curve will lead to certain peak values be submerged cause target focus lose
It loses.
Summary of the invention
The object of the present invention is to provide a kind of auto focusing methods, so that can be real to multiple objects in actual measurement
Now automatic to focus, the peak value at focusing is easier to differentiate.
The present invention adopts the following technical scheme that realization: a kind of automatic focusing side for the measurement of multiple target objects digital hologram
Method, comprising the following steps:
Step 1: including the hologram of multiple objects by the method acquisition that digital hologram records；
Step 2: the holographic images of each object are separated from collected hologram, separation include identification and
Two steps of label, the identification are complete to what is separated for accurately separating each holographic subgraph from hologram
Breath subgraph is focused automatically；The label is for corresponding to the relative position on holographic facet for each holographic subgraph；
Step 3: automatic seeking coke search range is set according to experiment condition, and step-length is set according to the range, will be isolated
The multiple holographic subgraphs come are rebuild by flag sequence seeking in burnt range, and each holography subgraph is in the step-length of setting, often
Step can obtain a width and rebuild picture；
Step 4: the reconstruction picture that every step obtains successively is calculated it with following equation and focuses criterion functional value, and is obtained certainly
Dynamic focusing curve,
Ci=exp [± k (Cri-<Cri>)]；
Wherein, Ih(i, j)=I (i, j)-<I>, Cri are that the image of every step focuses criterion functional value, and Ci is exponential transform
Criterion function is focused, automatic focusing curve can be obtained seeking in burnt range, I (i, j) indicates mesh of the current diffraction apart from lower reconstruction
Object image is marked, i, j indicate that image pixel indexes subscript, and<>expression takes mean value, indicates to take the equal of gray value of image in this formula
Value, IhFor the high fdrequency component of reproduction image,<I>is the gray value mean value of reproduction image；
Step 5: after all holographic subgraphs are handled by above-mentioned steps, using the automatic seeking coke search range of setting as cross
Coordinate, using Ci value as ordinate, the Ci Drawing of Curve that all holographic subgraphs are obtained, can according to the figure under unified coordinate system
Obtain the peak value of multiple objects, i.e., the specific reproduction distance of multiple objects；
Step 6: the specific reproduction distance of the multiple objects obtained according to step 5 rebuilds original hologram,
All reconstructions are obtained into the spatial distribution state of multiple target objects as being placed in uniform spaces coordinate system.
The invention has the following beneficial effects:
1, make the peak value at focusing more obvious by exponential transform automatic focusing criterion function, effectively reduce peak
Value distinguishes difficulty；
2, the transformed automatic focusing criterion function of utilization index successively calculates the hologram after segmentation, can avoid
Mutual crosstalk in object reproducing processes, the focal position of the clearly discernible each object out of the focusing curve of acquisition；
3, the present invention improves the reproduction distance discrimination speed and resolving ability in the measurement of digital hologram multiple target objects, thus
The accuracy and real-time for improving digital hologram measurement, are conducive to the further development of digital hologram target identification.
Detailed description of the invention
Fig. 1 is flow chart of the invention.
Fig. 2 is the hologram of the multiple objects of simulation of the embodiment of the present invention.
Fig. 3 is the automatic focusing curve figure obtained in the embodiment of the present invention using conventional method.
Fig. 4 is the automatic focusing curve figure obtained in the embodiment of the present invention using the present invention.
Specific embodiment
Present embodiment is illustrated below with reference to Fig. 1.Present embodiment, one kind are measured for multiple target objects digital hologram
Auto focusing method, comprising the following steps:
Step 1: including the hologram of multiple objects by the method acquisition that digital hologram records；
Step 2: the holographic images of each object are separated from collected hologram, separation include identification and
Two steps of label, the identification are complete to what is separated for accurately separating each holographic subgraph from hologram
Breath subgraph is focused automatically；The label is for corresponding to the relative position on holographic facet for each holographic subgraph；
Step 3: automatic seeking coke search range is set according to experiment condition, and step-length is set according to the range, will be isolated
The multiple holographic subgraphs come are rebuild by flag sequence seeking in burnt range, and each holography subgraph is in the step-length of setting, often
Step can obtain a width and rebuild picture；
Step 4: the reconstruction picture that every step obtains successively is calculated it with following equation and focuses criterion functional value, and is obtained certainly
Dynamic focusing curve,
Ci=exp [± k (Cri-<Cri>)]；
Wherein, Ih(i, j)=I (i, j)-<I>, Cri are that the image of every step focuses criterion functional value, and Ci is exponential transform
Criterion function is focused, automatic focusing curve can be obtained seeking in burnt range, I (i, j) indicates mesh of the current diffraction apart from lower reconstruction
Object image is marked, i, j indicate that image pixel indexes subscript, and<>expression takes mean value, indicates to take the equal of gray value of image in this formula
Value, IhFor the high fdrequency component of reproduction image,<I>is the gray value mean value of reproduction image；
Step 5: after all holographic subgraphs are handled by above-mentioned steps, using the automatic seeking coke search range of setting as cross
Coordinate, using Ci value as ordinate, the Ci Drawing of Curve that all holographic subgraphs are obtained, can according to the figure under unified coordinate system
Obtain the peak value of multiple objects, i.e., the specific reproduction distance of multiple objects；
Step 6: the specific reproduction distance of the multiple objects obtained according to step 5 rebuilds original hologram,
All reconstructions are obtained into the spatial distribution state of multiple target objects as being placed in uniform spaces coordinate system.
Cri in the present invention can be replaced with other focusing criterion function, be detailed in Sun Y, Duthaler S, Nelson B
J.Autofocusing in computer microscopy:Selecting the optimal focus algorithm
[J].Microscopy Research&Technique,2004,65(3):139-149。
Li Weiliang digital hologram focuses automatically and image quality Enhancement Technology [D].
Hologram of the method acquisition comprising multiple objects that the present invention is recorded by coaxial digital holography.In order to avoid certainly
Failure is focused in dynamic focusing caused by the mutual crosstalk of multiple objects, it need to be by the holographic images of each object from collecting
Hologram in separate.Separation process includes identification and two steps of label, the purpose of identification be by each holographic subgraph from
It is accurately separated in hologram, the holographic subgraph separated is focused automatically.The purpose of label is that each height is complete
Breath schemes the relative position on corresponding holographic facet, recovery when differentiating and reproduce convenient for the subsequent object focused automatically.
Hologram is first carried out grayscale equalization processing by identification process, and it is uneven to reduce the intensity profile as caused by experimental situation
The problem of, improve the efficiency of holographic subgraph separation.Treated, and hologram carries out binary conversion treatment again, highlights each object
Holographic images profile.Obtained binary image need to carry out image expansion processing, in the premise for not influencing to separate between object
Under the degree of image expansion can be improved, guarantee the integrality of each sub-hologram of separation.The subgraph isolated will appear image
Problem not of uniform size, therefore zero padding processing need to be carried out, by the position zero padding other than subgraph pixel coverage after setting image size
Filling is so as to program processing.Labeling process need to be by the corresponding mark in order of each object and the subgraph separated in hologram
Note, it is therefore an objective to which data save and obtain position distribution on the face of object when reproducing with the later period when facilitating processing.
Automatic seeking coke search range is set according to experiment condition, each holographic subgraph separated successively is existed by label
It seeks in burnt range and being rebuild, reconstruction process, which angular spectrum method can be used to obtain reproduction image, can also be used other digital hologram reproducting methods
It is reproduced, every step will obtain a width and rebuild picture in the step-length of setting, and every step is rebuild as successively calculating it with following equation
Focus criterion functional value:
Ci=exp [± k (Cri-<Cri>)]
The step utilization index transformed focusing criterion function obtains the automatic focusing curve in the case where setting step-length, wherein
Ih(i, j)=I (i, j)-<I>, I (i, j) indicate current diffraction apart from lower reconstruction
Target object image, i, j indicate image pixel index subscript,<>expression take mean value, in this formula indicate take gray value of image
Mean value.
± value is related with measured target object in Ci expression formula, when measurement target is that infiltration type object takes negative sign, non-transmission
Type object takes positive sign；K is affected to the peak value of Ci, and k value is related with the range of Cri, need to judge the variation range of Cri, if
Cri starts to change for n-th after decimal point, then value k >=10n, apparent criterion function curve can be obtained in the value for meeting k
Peak value, k takes 10 in example of the present invention2.Wherein the form of Cri uses the standard deviation criterion of image grayscale high fdrequency component, can also be with
For expression-forms such as entropy criterion, gray variance criterion or shade of gray criterions.The step is meant script peak value under multiple target objects not
Apparent focus function value greatly increases the discrimination degree of peak value by exponential transform.
Each subgraph after separation successively carries out upper one step process according to the sequence marked and calculates focus function curve, saves
The Ci value obtained after per treatment sits the burnt range of seeking of setting as abscissa after the processing of all subgraphs using Ci value as vertical
Mark, then the Ci Drawing of Curve that all subgraphs are obtained can quickly be read multiple under the same coordinate system according to total curve graph
The specific reproduction distance of object.
After the reproduction distance of each object obtained by upper step, successively substitute into each reproductions distance use angular spectrum reproducting method or
Other digital hologram reproducting methods rebuild original hologram, can obviously observe object by the image after rebuilding
Picture, all target reproduction images are placed in the same coordinate system and obtain the spatial distribution state of multiple target objects.
The form of Cri may be entropy criterion, gray variance criterion or shade of gray criterion etc. in Ci formula in the present invention
Expression-form.The step is meant makes the discrimination degree of peak value increase the not high focus function value of script peak value by exponential transform
Add.
Embodiment:
The present invention is discussed in detail below with reference to Fig. 2 to Fig. 4.It should be noted that embodiment described herein only to
It explains the present invention, is not intended to limit the present invention.As long as in addition, technical characteristic involved by invention described below that
Not constituting conflict around here can be combined with each other.
Cross transparent objects object is chosen in emulation, object is altogether there are four A, B, C, D is respectively labeled as, with receiving plane
Distance is set as 10cm, 11cm, 12cm and 13cm, and the hologram being calculated is as shown in Figure 2.By shown in Fig. 2 comprising all
The hologram of object obtains focusing curve figure as shown in Figure 3 after being calculated with traditional auto-focusing algorithm, object in figure
Specific location is not easy to find, adds the difficulty of discrimination, affect the speed of holographic measurement middle and later periods processing.
The present invention splits each object to obtain four holographic subgraphs, each subgraph according to fig. 2 in the hologram
A corresponding object, and by each subpicture tag, judge the Cri fluctuation range of each subgraph, after giving reasonable k,
K takes 10 in example of the present invention2, the automatic focusing curve of each subgraph is calculated with Ci formula, and all focusing curves are drawn
In a coordinate system, as shown in figure 4, Fig. 4 comparison diagram 3 finds that the reproduction distance of four targets is clearly easily debated, be conducive to improve
The efficiency and accuracy of multiple target objects digital hologram measurement.
It should be noted that specific embodiment is only the explanation and illustration to technical solution of the present invention, it cannot be with this
Limit rights protection scope.What all claims according to the present invention and specification were made is only locally to change, Reng Yingluo
Enter in protection scope of the present invention.
Claims (7)
1. a kind of auto focusing method for the measurement of multiple target objects digital hologram, it is characterised in that the following steps are included:
Step 1: including the hologram of multiple objects by the method acquisition that digital hologram records；
Step 2: the holographic images of each object are separated from collected hologram, and separation includes identification and label
Two steps, the identification from hologram for accurately separating each holographic subgraph, to holography separated
Figure is focused automatically；The label is for corresponding to the relative position on holographic facet for each holographic subgraph；
Step 3: automatic seeking coke search range is set according to experiment condition, and step-length is set according to the range, by what is separated
Multiple holography subgraphs are rebuild by flag sequence seeking in burnt range, and each holography subgraph is in the step-length of setting, every step meeting
It obtains a width and rebuilds picture；
Step 4: the reconstruction picture that every step obtains successively is calculated it with following equation and focuses criterion functional value, and is gathered automatically
Burnt curve,
Ci=exp [± k (Cri-<Cri>)]；
Wherein, Ih(i, j)=I (i, j)-<I>, Cri are that the image of every step focuses criterion functional value, and Ci is the focusing of exponential transform
Automatic focusing curve can be obtained seeking in burnt range in criterion function, and I (i, j) indicates object of the current diffraction apart from lower reconstruction
Image, i, j indicate that image pixel indexes subscript,Expression takes mean value, and the mean value for taking gray value of image, I are indicated in this formulahFor
The high fdrequency component of reproduction image,<I>are the gray value mean value of reproduction image；
Step 5: by all holographic subgraphs by above-mentioned steps processing after, using the automatic seeking coke search range of setting as abscissa,
Using Ci value as ordinate, the Ci Drawing of Curve that all holographic subgraphs are obtained is available according to the figure under unified coordinate system
The peak value of multiple objects, i.e., the specific reproduction distance of multiple objects；
Step 6: the specific reproduction distance of the multiple objects obtained according to step 5 rebuilds original hologram, by institute
There is reconstruction to obtain the spatial distribution state of multiple target objects as being placed in uniform spaces coordinate system.
2. a kind of auto focusing method for the measurement of multiple target objects digital hologram according to claim 1, feature exist
In the detailed process identified in the step 2 are as follows: hologram is successively carried out to grayscale equalization processing, binary conversion treatment, swollen
Swollen processing and zero padding processing.
3. a kind of auto focusing method for the measurement of multiple target objects digital hologram according to claim 1, feature exist
In the detailed process marked in the step 2 are as follows: by multiple objects and the holographic subgraph separated in hologram by
Sequence correspondence markings.
4. a kind of auto focusing method for the measurement of multiple target objects digital hologram according to claim 1, feature exist
In: the method for the reconstruction is that digital hologram reproduces method.
5. a kind of auto focusing method for the measurement of multiple target objects digital hologram according to claim 4, feature exist
In: it is angular spectrum method that the digital hologram, which reproduces method,.
6. a kind of auto focusing method for the measurement of multiple target objects digital hologram according to claim 1, feature exist
In the alternative forms for focusing criterion function Cri in the step 4 are entropy criterion function, gray variance criterion function or gray scale ladder
Spend criterion function.
7. a kind of auto focusing method for the measurement of multiple target objects digital hologram according to claim 1, feature exist
In the step 1 uses coaxial digital holography writing-method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910244991.0A CN109828444A (en) | 2019-03-28 | 2019-03-28 | A kind of auto focusing method for the measurement of multiple target objects digital hologram |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910244991.0A CN109828444A (en) | 2019-03-28 | 2019-03-28 | A kind of auto focusing method for the measurement of multiple target objects digital hologram |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109828444A true CN109828444A (en) | 2019-05-31 |
Family
ID=66873167
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910244991.0A Pending CN109828444A (en) | 2019-03-28 | 2019-03-28 | A kind of auto focusing method for the measurement of multiple target objects digital hologram |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109828444A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110286464A (en) * | 2019-06-14 | 2019-09-27 | 浙江大学 | A kind of auto focusing method based on Area rule |
| CN110675383A (en) * | 2019-09-24 | 2020-01-10 | 哈尔滨工业大学 | Target object rapid identification focusing method for interference object in digital holography |
| CN112882370A (en) * | 2021-01-21 | 2021-06-01 | 南京邮电大学 | Method for obtaining accurate reconstruction distance during automatic focusing of digital holographic imaging |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103914870A (en) * | 2014-02-28 | 2014-07-09 | 天津工业大学 | High-universality automatic hologram reestablishing method based on new focus evaluation function |
| CN105242268A (en) * | 2015-10-09 | 2016-01-13 | 中国电子科技集团公司第五十研究所 | Automatic focusing imaging method for short-range holographic radar |
| CN106228523A (en) * | 2016-07-28 | 2016-12-14 | 新疆大学 | A kind of method improving noisy digital hologram quality reproduction |
| CN106502074A (en) * | 2016-11-08 | 2017-03-15 | 华中科技大学 | A kind of auto focusing method for image planes digital holographic micro-measuring |
| CN107966111A (en) * | 2017-11-21 | 2018-04-27 | 北京理工大学 | A kind of Atomatic focusing method for digital hologram restructuring procedure |
-
2019
- 2019-03-28 CN CN201910244991.0A patent/CN109828444A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103914870A (en) * | 2014-02-28 | 2014-07-09 | 天津工业大学 | High-universality automatic hologram reestablishing method based on new focus evaluation function |
| CN105242268A (en) * | 2015-10-09 | 2016-01-13 | 中国电子科技集团公司第五十研究所 | Automatic focusing imaging method for short-range holographic radar |
| CN106228523A (en) * | 2016-07-28 | 2016-12-14 | 新疆大学 | A kind of method improving noisy digital hologram quality reproduction |
| CN106502074A (en) * | 2016-11-08 | 2017-03-15 | 华中科技大学 | A kind of auto focusing method for image planes digital holographic micro-measuring |
| CN107966111A (en) * | 2017-11-21 | 2018-04-27 | 北京理工大学 | A kind of Atomatic focusing method for digital hologram restructuring procedure |
Non-Patent Citations (6)
| Title |
|---|
| HAZAR A. ILHAN, MERTDO˘GAR, MERIC OZCAN: "Fast autofocusing in digital holography using scaled holograms", 《OPTICS COMMUNICATIONS》 * |
| YU SUN,STEFAN DUTHALER,BRADLEY J.NELSON: "Autofocusing in computer microscopy:Selecting the Optimal Focus Algorithm", 《MICROSCOPY RESEARCH AND TECHNIQUE》 * |
| 张宇，侯春风，孙秀冬: "Grey spatial solitons due to two-photon photorefractive effect", 《CHINESE PHYSICS》 * |
| 徐元强，王玉荣，海生: "数字同轴全息颗粒场检测中的颗粒分割识别与独立聚焦", 《光子学报》 * |
| 李国: "基于遗传算法的遥感影像增强技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 李维亮: "数字全息自动聚焦及像质增强技术研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110286464A (en) * | 2019-06-14 | 2019-09-27 | 浙江大学 | A kind of auto focusing method based on Area rule |
| CN110675383A (en) * | 2019-09-24 | 2020-01-10 | 哈尔滨工业大学 | Target object rapid identification focusing method for interference object in digital holography |
| CN110675383B (en) * | 2019-09-24 | 2023-04-14 | 哈尔滨工业大学 | Target object rapid identification focusing method for interference object in digital holography |
| CN112882370A (en) * | 2021-01-21 | 2021-06-01 | 南京邮电大学 | Method for obtaining accurate reconstruction distance during automatic focusing of digital holographic imaging |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109828444A (en) | A kind of auto focusing method for the measurement of multiple target objects digital hologram | |
| Feng et al. | Fringe pattern analysis using deep learning | |
| CN103201769A (en) | Image processing device, image processing method, program, integrated circuit | |
| KR101382892B1 (en) | Method of recognizing low-resolution image face and low resolution image face recognition device | |
| CN106227015A (en) | A kind of hologram image super-resolution reconstruction method and system based on compressive sensing theory | |
| CN112529791A (en) | Adaptive multifocal restoration method based on plankton digital holographic image | |
| Anand et al. | Digital holographic microscopy for automated 3D cell identification: an overview | |
| CN115330846A (en) | Three-dimensional digital holography-based pavement disease image fusion data processing method | |
| Galande et al. | Quantitative phase imaging of biological cells using lensless inline holographic microscopy through sparsity-assisted iterative phase retrieval algorithm | |
| Patel et al. | Application of image-to-image translation in improving pedestrian detection | |
| CN109636860A (en) | Illumination patterns estimation and replay method and the device of image | |
| CN104751412B (en) | A kind of image split-joint method based on affine invariants | |
| CN116912138A (en) | Dynamic multi-exposure light field image fusion method based on structure consistency detection | |
| CN113838186B (en) | Fresnel hologram frequency domain gating filtering method based on deep learning | |
| Gong et al. | Crude Oil Leakage Detection Based on DA‐SR Framework | |
| CN110286464B (en) | Automatic focusing method based on area criterion | |
| Phillipson et al. | Old School’: An 8K Multicamera Shoot to Create a Dataset for Computational Cinematography | |
| Yi et al. | Extraction of target specimens from bioholographic images using interactive graph cuts | |
| JP7378164B2 (en) | Optical system and optical image processing method applying image restoration | |
| Liu et al. | Video forgery detection using multiple cues on fusion of efficientnet and swin transformer | |
| Zheng et al. | Edge detection ghost imaging using frequency-domain filtering | |
| Nathiya et al. | GAN with CCSO: generative adversarial network-driven CAViaR competitive swarm optimizer for medical video super resolution | |
| CN110610470B (en) | Camera multi-focus clear image extraction method based on multi-azimuth gradient comparison | |
| CN117058049B (en) | New view image synthesis method, synthesis model training method and storage medium | |
| Pena et al. | A Low-Cost Process for Plant Motion Magnification for Smart Indoor Farming. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190531 |