CN112116675B - Ghost imaging optimization method based on multi-speckle pattern combination-modulation - Google Patents
Ghost imaging optimization method based on multi-speckle pattern combination-modulation Download PDFInfo
- Publication number
- CN112116675B CN112116675B CN202010923096.4A CN202010923096A CN112116675B CN 112116675 B CN112116675 B CN 112116675B CN 202010923096 A CN202010923096 A CN 202010923096A CN 112116675 B CN112116675 B CN 112116675B
- Authority
- CN
- China
- Prior art keywords
- modulation
- speckle pattern
- ghost
- speckle
- ghost imaging
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
- G06T11/005—Specific pre-processing for tomographic reconstruction, e.g. calibration, source positioning, rebinning, scatter correction, retrospective gating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
Abstract
The invention relates to ghost imaging optimization, in particular to a ghost imaging optimization method based on multi-speckle pattern combination-modulation, which sequences a plurality of speckle patterns irradiated on a target object by using the value of a bucket detector value so as to reduce the difference between adjacent speckle patterns; through the superposition of adjacent speckle patterns and the modulation of corresponding barrel detector values, the redundancy and the quantity of data to be associated are effectively reduced, and the reconstruction of a target object is realized by utilizing different association operation rules. Numerical simulation results show that for a target image G, under the combination mode of 4000 times of total sampling times and 4 adjacent speckle patterns, the peak signal-to-noise ratio/contrast of the provided method and the traditional ghost imaging, differential ghost imaging and positive-negative modulation ghost imaging under no combination-modulation are respectively improved by 21.7%/27.3%, 8.3%/17.8% and 14.7%/25.7%; through numerical simulation and result analysis of 20 target images, the ratios of the peak signal-to-noise ratio/the contrast improvement rate of more than 15% and 30% are respectively 90%/85% and 50%/55%, and the method has good generalization capability.
Description
Technical Field
The invention relates to optimization of ghost imaging, in particular to a ghost imaging optimization method based on multi-speckle pattern combination-modulation.
Background
The ghost imaging technology is an indirect imaging method for reconstructing target object information by using a spatial intensity correlation principle, can break through the limitation that array detectors are difficult to respond due to insufficient light energy in traditional imaging, has the unique advantages of strong anti-interference performance, long action distance, all-weather monitoring and the like, and has wide application prospects in the fields of night city security and monitoring, industrial/medical ultralow-radiation CT images and the like. Compared with the traditional imaging mode, the detection process and the reconstruction process of the ghost imaging technology are completely separated, wherein the detection process mainly obtains the measured value of the bucket detector by changing the speckle pattern irradiated on the target area, and the reconstruction realizes the indirect reconstruction of the target ghost image by the correlation operation between the speckle pattern and the measured value.
In the ghost imaging process, there must in principle be a significant difference between each individual speckle pattern illuminated on the illuminated object, only to obtain a high quality ghost image of the target object in the non-object path. For example: a ghost imaging method based on Hadamard derivative images is mainly characterized in that a Hadamard measurement matrix is used for generating a series of speckle images so as to reduce redundancy among the speckle images. In 2015, Mahdi Khamoushi et al proposed and verified operability of Sinusoidal Ghost Imaging (SGI) by frequency domain decomposition of fourier series. A computed ghost imaging method based on orthogonal sinusoidal speckles is provided by subject group personnel in a mode of superposing two orthogonal oblique sinusoidal speckle patterns. The methods are based on the orthogonal idea to increase the difference between speckle patterns, and theoretically, a target object can be completely reconstructed on the premise of full sampling, which shows that the ghost imaging quality can be effectively improved by the large difference. However, in practical detection, due to the influence of the practical detection environment at night and the limitation of the ultra-low radiation modulator, the distribution of speckle patterns irradiated on the target object is often random rather than orthogonal, so that the difference between adjacent speckle patterns is difficult to be highlighted, thereby causing that the spatial information of the target object is difficult to be effectively mined during the correlation reconstruction, and the imaging effect is poor.
Disclosure of Invention
In view of the above analysis, the present application proposes a multi-speckle pattern combination-modulation method that facilitates highlighting differences between associated data by using a combination of multiple speckle patterns and modulation transformation between corresponding bucket detector values on the premise of a random speckle pattern, and uses the multi-speckle pattern combination-modulation method in optimization of ghost imaging quality. The method of the invention improves the difference significance between the adjacent speckle patterns by sequencing and combining a plurality of speckle patterns; the modulation of the corresponding bucket detector values is used to effectively enhance the inter-data differences. Through numerical simulation, a setting method of the number of speckle pattern combinations is analyzed, and effectiveness and generalization of the method are verified by utilizing quantitative evaluation indexes such as peak signal-to-noise ratio and contrast.
The invention is realized by adopting the following technical scheme: a ghost imaging optimization method based on multi-speckle pattern combination-modulation comprises the following steps: the method comprises the following steps:
(1) based on measured bucket detector value I bm The sizes of the sequences are sequenced from small to big to obtain a sequence { I' b1 ,I′ b2 ,L,I′ bK And in this sequence { I' b1 ,I′ b2 ,L,I′ bK According to it, the value of the bucket detector I bm Corresponding speckle pattern I am (x, y) are also rearranged to give the sequence { I' a1 ,I′ a2 ,L,I′ aK -reducing the difference between adjacent speckle patterns;
(2) sequence { I 'of sorted speckle patterns' a1 ,I′ a2 ,L,I′ aK Sequentially overlapping and combining the adjacent speckle patterns to form a new speckle pattern sequence
Reducing redundancy and quantity among data to be associated; the ith newly generated speckle pattern is:
wherein i is 1,2, L, K/L, and K is an integral multiple of L;
(3) multiplicative modulation is carried out on the barrel detector values corresponding to the superposed and combined one-piece speckle pattern to obtain a new barrel detector sequence value
Then the ith newly generated bucket detector value is:
(4) utilizing ghost imaging method to obtain new speckle pattern sequence
Sum bucket detector value sequence
And performing correlation operation to obtain a ghost image.
The ghost imaging optimization method based on multi-speckle pattern combination-modulation belongs to [2,8 ].
The technical principle is that the difference degree between data is increased by preprocessing the speckle patterns and the bucket detector values (namely, the combination between the multiple speckle patterns and the modulation of the bucket detector values), so that the redundancy between the associated data is reduced, and the imaging quality is improved. This approach is equivalent to adding a pre-processing procedure to the detection system data (i.e., speckle pattern, bucket detector values) prior to correlation, resulting in increased differences between the correlated data. The difference between speckle patterns is an important factor for determining the quality of ghost imaging, and the improvement of the difference between random speckle patterns in the correlation operation is an effective way for effectively improving the peak signal-to-noise ratio, the contrast ratio and the like of ghost images. Aiming at the problem, the invention provides a novel ghost imaging optimization method by utilizing a mode of combining a plurality of speckle patterns. The method comprises the steps of firstly, sequencing a plurality of speckle patterns irradiated on a target object by using the value of a barrel detector, so that the difference between the adjacent speckle patterns is reduced; through the superposition of adjacent speckle patterns and the modulation of corresponding barrel detector values, the redundancy and the quantity of data to be associated are effectively reduced, and the reconstruction of a target object is realized by utilizing different association operation rules. The numerical simulation result shows that for a target image G, under the combination mode of 4000 times of total sampling times and 4 adjacent speckle patterns, the peak signal-to-noise ratio/contrast of the traditional ghost imaging, the differential ghost imaging and the positive and negative modulation ghost imaging under the condition of no combination-modulation is respectively improved by 21.7%/27.3%, 8.3%/17.8% and 14.7%/25.7%; through numerical simulation and result analysis of 20 target images, the ratio of peak signal-to-noise ratio/contrast improvement rate respectively above 15% and 30% is 90%/85% and 50%/55%, which shows that the method has good generalization capability. The method can be popularized in all the existing correlation algorithms, and the imaging effect is obviously improved compared with the prior art.
Drawings
Fig. 1 is a speckle pattern and intensity distribution of a gaussian speckle beam.
FIG. 2 is a schematic view of the process of the present invention.
FIG. 3 is a graph of numerical simulation results of different methods.
FIG. 4 is a contrast plot of reconstructed ghost images at different combined lengths.
FIG. 5 is a graph of peak signal-to-noise ratios of reconstructed ghost images at different combined lengths.
FIG. 6 is a graph of contrast and peak SNR for reconstructed ghost images of different target objects and different combined lengths
FIG. 7 is a graph of the results of numerical simulations for different combination lengths.
FIG. 8 is a comparison of the multi-speckle pattern combination-pre-and post-modulation ghost results.
Fig. 9 is a graph showing peak signal-to-noise ratio and contrast boost amplitude.
Detailed Description
Proposal of multiple speckle pattern combination problem
Setting the mth random speckle pattern for illuminating the target object to be I under the random speckle field environment am (x, y) (where m is 1,2, L, K), and the bucket detector value after the object is illuminated is I bm Then, the target object may be reconstructed by using a conventional second order correlated ghost imaging (TGI), Differential Ghost Imaging (DGI), and positive-negative ghost imaging (P-NGI) method, which specifically includes:
wherein G' is a reconstructed ghost image; g' is positive when the + is positive or negative when the + is negative; f (g) is a modulation function, which may be generally a function of logarithm, power exponent, positive (cosine), etc.
To measure the similarity between random speckle patterns, we will use 10 sparse gaussian speckle beams (as shown in fig. 1) as an example. The size of each speckle pattern is 64 multiplied by 64, the position points of the speckle beams in each speckle pattern are random, and the intensity of the speckle beams is subjected to Gaussian distribution.
When the number of the speckle patterns is K, the similarity between the speckle patterns can be described by using the distance. The process mainly comprises two parts: first, according to the value of the bucket detector I bm (m ═ 1,2, L, K) for their corresponding speckle patterns I in descending order am (x, y) sorting; secondly, the Euclidean distance is utilized to calculate the distance between any two adjacent speckle patterns after sequencing, and an average distance matrix is obtained. When K is 4000, the average distance matrix is:
summing all elements of the average distance matrix and averaging
This means that the distance between adjacent random speckle patterns is relatively small, i.e. the similarity between speckle patterns is high. Therefore, the ghost quality can only be improved by increasing the sampling times, but this will tend to increase the detection and reconstruction time.
In principle, when the sampling times are constant, the ghost image quality can also be improved by increasing the difference between speckle patterns. Therefore, in order to reduce the similarity between random speckle patterns, the present invention transforms adjacent speckle patterns in the form of a superposition combination. The description is given by taking the superposition combination of 4 adjacent speckle patterns as an example, and the combined average distance matrix is:
summing all elements of the average distance matrix and averaging
As can be seen from the above, after combining the adjacent speckle patterns, the difference between the speckle patterns is increased by 3.05 times compared with that before combining, which is certainly beneficial to the improvement of ghost imaging quality. However, as the number of adjacent speckle pattern combinations increases, the number of new speckle patterns after combination tends to decrease, and when the number of new speckle patterns decreases to a certain extent, it is not favorable for imaging in principle. That is, the larger the number of combinations of adjacent speckle patterns is, the better.
Ghost imaging optimization method of multi-speckle pattern combination-modulation
Concrete implementation process of method
The principle schematic diagram of the multi-speckle pattern combination-modulated ghost imaging optimization method is shown in fig. 2. Wherein, the series of random speckle patterns are irradiated on a Spatial Light Modulator (SLM) by laser to simulate a light field under an actual detection environment; I.C. A a1 ,I a2 ,L,I aK Speckle pattern for the original illuminated target object, I b1 ,I b2 ,L,I bK Is the corresponding bucket detector value; i' b1 ,I′ b2 ,L,I′ bK Is the sequence of the bucket detector values which are sequenced from small to big, and the corresponding speckle pattern sequence is I' a1 ,I′ a2 ,L,I′ aK (ii) a l is the combined number of adjacent speckle patterns, called as the combined length, and in general, K is an integral multiple of l;
respectively, the combined-modulated bucket detector values and speckle pattern sequences.
The implementation process of fig. 2 mainly includes four parts of sorting, combining, modulating, and associating, specifically:
(1) based on measured bucket detector value I bm Is sorted from small to large by { I' b1 ,I′ b2 ,L,I′ bK Based on this, the corresponding speckle pattern I is checked am (x, y) is also rearranged { I' a1 ,I′ a2 ,L,I′ aK Is used to reduce the difference between adjacent speckle patternsLow.
(2) The ordered speckle pattern sequence is overlapped and combined with the adjacent speckle patterns according to the sequence from left to right to form a new speckle pattern sequence
The redundancy and the amount of data to be associated are reduced. The ith newly generated speckle pattern is:
wherein, i is 1,2, L, K/L.
(3) Multiplicative modulation is carried out on the barrel detector values corresponding to the superposed and combined one-piece speckle pattern to obtain a new barrel detector sequence value
Then the ith newly generated bucket detector value is:
(4) new speckle pattern sequence obtained by TGI, DGI and P-NGI ghost imaging method
Sum bucket detector value sequence
And performing correlation operation to obtain a ghost image.
Wherein, G' l Reconstructing ghost images after combined modulation; g' is positive when the + is positive or negative when the + is negative; f (g) is a modulation function, which may be generally a function of logarithm, power exponent, positive (cosine), etc.
Specifically, when the combined length l is 1, which is equivalent to not performing the combination-modulation on the speckle pattern, equation (4) is equivalent to equation (1); when the combination length is equal to the total sampling number, i.e., l ═ K, it is equivalent to combining all speckle patterns together, and the correlation reconstruction cannot be performed at this time. It can be seen that, as the combined length l increases, the degree of improvement of the ghost quality does not gradually increase, but shows a trend of increasing first and then decreasing. Since the combined length l has some influence on the quality of the reconstructed ghost image, it was found through research that it is usually constant and l ∈ [2,8 ].
Evaluation index
In order to more vividly explain the imaging quality of the method, the invention utilizes two evaluation indexes of peak signal-to-noise ratio and contrast to carry out quantitative evaluation on the reconstructed ghost image, and the method specifically comprises the following steps:
(1) peak signal to noise ratio
Wherein, the first and the second end of the pipe are connected with each other,
g (i, j) is the transmittance of each pixel point, and x and y are the sizes of the reconstructed images;
is the maximum gray value in the ghost image.
(2) Contrast ratio
The contrast v in ghost imaging is defined with reference to the Michelson contrast, i.e.
Wherein G is P And G V Respectively, the peak value and the valley value on the reconstructed ghost image gray value curve.
Numerical simulation and analysis
To testTo demonstrate the effectiveness of the proposed method, the target object is illuminated with 4000 random speckle patterns of 64 x 64 size, and is modulated by TGI, DGI and f (g) log before combination with no modulation 2 (g) And the P-NGI result is subjected to comparative analysis to illustrate the advantages of the method in improving the ghost image quality.
(1) Reconstructing a particular target image
Taking an unknown target image "G" as an example, different ghost imaging methods are used to respectively reconstruct target images before combining-modulating and after combining-modulating, and a gray value curve of a ghost image in a certain dimension is plotted, as shown in fig. 3, in this experiment, the combining step length l is 4. The purpose of the gray value curve here is to analyze and calculate the contrast of different ghost image results. In FIG. 3, (a) to (d) are respectively the target image to be reconstructed, the TGI reconstruction result, the DGI reconstruction result, and the log-based data 2 (g) The modulated P-NGI positive image reconstruction result; (f) distribution graphs of gray values of transposed 31 st row of pixels corresponding to ghost images corresponding to the three methods after reconstruction are respectively obtained; (i) when the object to be measured and the combined length are 4, the invention combines TGI and DGI respectively and is based on log 2 (g) A reconstructed ghost result after the modulated P-NGI method; and (m) to (p) are distribution graphs of gray values of transposed pixels of the 31 st column corresponding to the ghost images of (i) to (l).
As can be seen from fig. 3, the reconstruction result of the method of the present invention can clearly distinguish the object region from the background region. The peak signal-to-noise ratios of the ghost images in the (b) to the (d) are respectively calculated as 10.9751, 13.5252 and 11.9594 by using the formula (6), and the peak signal-to-noise ratios of the ghost images in the (j) to the (l) after the multi-speckle pattern combination-modulation are respectively 13.3567, 14.6478 and 13.7225. Therefore, the peak signal-to-noise ratios obtained by combining and modulating the speckle patterns and then utilizing different ghost imaging methods are improved, and the improvement degrees are respectively 21.7%, 8.3% and 14.7%. Calculating the contrast of the target image and the ghost images in the (b) to (d) before the combination and modulation by using the formula (5) according to the gray value curves (e) to (h) to be 1, 0.33, 0.45 and 0.35 respectively; from the gray-scale value curves (n) - (p), the contrast of the ghost image in (j) - (l) is improved to 0.42, 0.53 and 0.44 respectively after the multi-speckle pattern combination-modulation. The comparison shows that the contrast of the speckle pattern after combination-modulation is improved by 27.3 percent, 17.8 percent and 25.7 percent respectively compared with that before combination-modulation.
The combined length l used in the above experiment is 4, but actually, when the values of the combined length are different, differences occur between reconstructed ghost images. In order to facilitate reasonable setting of the combination length, the contrast and the signal-to-noise ratio of the reconstructed ghost images at different combination lengths are analyzed below, and corresponding graphs are respectively shown in fig. 4 and 5. Wherein the abscissa is the combination length, the ordinate is the calculated contrast, and graph (b) is a graph obtained by transforming the abscissa into a logarithmic representation on the basis of graph (a), in order to give more detail the correspondence between the curve in graph (a) and the combination length l at the rise phase.
When log is shown in FIG. 4 and FIG. 5, it can be seen that 2 (l) The value range is [1,3 ]]When the contrast and peak signal-to-noise ratio curves of the reconstructed image show continuously rising phases, and when the log is recorded 2 (l) And when the peak signal-to-noise ratio is more than or equal to 4, the peak signal-to-noise ratio shows a descending trend. Meanwhile, the data volume of the data to be reconstructed is closely related to the imaging quality, and the imaging quality is better and better along with the increase of the data volume, so that the log is equivalent to 2 (l) When the value is larger than or equal to 4, the smaller the data volume of the data to be reconstructed, the less the improvement of the imaging quality is, namely, l is not larger, and the better l is. Therefore, when log 2 (l)∈[1,3]I.e. l ∈ [2,8]]In the process, the contrast ratio and the peak signal-to-noise ratio of the ghost image are both high, and the data volume of the data to be reconstructed is also large, so that the imaging is facilitated.
From the overall trend of the curves in fig. 4 and 5, when the combined length l is increased gradually, the contrast and peak signal-to-noise ratio curves both show a trend of increasing and decreasing, and the interval ranges of the maximum values are approximately the same. This shows that under the same appropriate combined length, the peak signal-to-noise ratio and contrast of the ghost image are both large, and the trade-off between the two is not needed to be balanced.
To illustrate the applicability of the variation trends of fig. 4 and 5 to other target objects, 5 different target objects are taken as an example, and the contrast-to-peak signal-to-noise ratio curves of the reconstructed ghost images at different combination lengths are shown in fig. 6.
As can be seen from fig. 6, although the contrast and the peak snr curves of the target ghost images at different combination lengths are different for different target objects, the overall variation trend is similar, and both the variation trend is increased and then decreased, which is consistent with the previous theoretical analysis conclusion.
In order to facilitate selection and setting of the combination length l, 20 different target images to be reconstructed are taken as reconstruction objects, reconstructed ghost images in a region (i ranges from 2 to 32) near a peak value are analyzed, and correlation operation is performed on speckle patterns and a bucket detector under different combination lengths by using TGI (trigloss tangent) by taking different combination lengths of 2, 4, 8, 16, 32 and the like as examples. Still taking the target image "G" as an example, when l is 2, 4, 8, 16 and 32, respectively, the ghost and the distribution of the gray-scale values of the reconstructed target image "G" are shown in fig. 7(a) to (e) and (f) to (j), respectively.
In fig. 7, (f) to (j) are distribution diagrams of the grayscale values of the 31 st column of the ghost images (a) to (e) after the transposition. It is obvious from the figure that the contrast of the reconstructed image in the range is in a direct proportion relation with the modulation step length, the contrasts obtained by calculation according to (f) to (j) are respectively 0.37, 0.42, 0.58, 0.70 and 0.71, the whole body shows a gradually increasing trend, and the increasing amplitude is increased and reduced firstly; the peak snr of graphs (a) to (e) is 11.8739, 13.3567, 13.7844, 13.4806 and 13.4362, respectively, and as a whole, shows a trend of increasing first and then decreasing. In particular, when l is 8, 16 and 32, the structure of the reconstructed ghost image is not complete, a plurality of connected regions are fractured, and the fracture phenomenon is more obvious along with the increase of l. In addition, in order to illustrate the selection basis of l under different sizes, the method is used for performing combination-modulation reconstruction on target images such as 30 × 30 and 128 × 128 images, and as a result, the optimal setting range of l is selected from 2 to 8 when the sampling times of speckles under different sizes are the same with the size ratio of the sizes.
(2) General applicability analysis of the proposed method
Since the above results are analyzed for a specific target image, the effectiveness of the proposed method for reconstructing other target images cannot be explained. Therefore, in order to verify the generalization of the method, 20 different target images to be reconstructed are used as reconstruction objects, the TGI is used to reconstruct the ghost images before and after the multi-speckle pattern combination-modulation, and the peak signal-to-noise ratio and the contrast corresponding to each reconstruction result are calculated, as shown in fig. 8.
Comparing the results in fig. 8, it is found that the peak signal-to-noise ratio and the contrast of the ghost image obtained by combining and modulating the speckle pattern are both improved, and the improvement amplitude is as shown in fig. 9.
As can be seen from fig. 9, when the target image in the target image set to be reconstructed is reconstructed by using the method of the present invention, the peak signal-to-noise ratio and the contrast of the ghost image are both improved to different degrees, the ratio of the peak signal-to-noise ratio and the contrast which are both improved by more than 15%/30% is 85%/50%, the improvement degree of the contrast is slightly higher than the peak signal-to-noise ratio, and the ratio of the improvement degree which is more than 15%/30% can reach 90%/55%. Therefore, the method has better universality and popularization value in the aspect of improving the quality of the reconstructed target.
Claims (2)
1. A ghost imaging optimization method based on multi-speckle pattern combination-modulation is characterized by comprising the following steps: the method comprises the following steps:
(1) based on measured bucket detector value I bm The sizes of the sequences are sequenced from small to big to obtain a sequence { I' b1 ,I′ b2 ,…,I′ bK And in this sequence { I' b1 ,I′ b2 ,…,I′ bK According to it, the value of the bucket detector I bm Corresponding speckle pattern I am (x, y) is also rearranged to obtain a sequence;
(2) sequence { I 'of sorted speckle patterns' a1 ,I′ a2 ,…,I′ aK Sequentially overlapping and combining the adjacent speckle patterns to form a new speckle pattern sequence
Reducing redundancy and quantity among data to be associated; the ith newly generated speckle pattern is:
wherein, i is 1,2, …, K/l;
(3) multiplicative modulation is carried out on the barrel detector values corresponding to the superposed and combined one-piece speckle pattern to obtain a new barrel detector sequence value
Then the ith newly generated bucket detector value is:
(4) new speckle pattern sequence obtained by ghost imaging method
And bucket detector value sequence
And performing correlation operation to obtain a ghost image.
2. A method for ghost imaging optimization based on multi-speckle pattern combining-modulation according to claim 1, characterized by: l is epsilon [2,8 ].
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010923096.4A CN112116675B (en) | 2020-09-04 | 2020-09-04 | Ghost imaging optimization method based on multi-speckle pattern combination-modulation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010923096.4A CN112116675B (en) | 2020-09-04 | 2020-09-04 | Ghost imaging optimization method based on multi-speckle pattern combination-modulation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112116675A CN112116675A (en) | 2020-12-22 |
| CN112116675B true CN112116675B (en) | 2022-09-23 |
Family
ID=73802259
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010923096.4A Active CN112116675B (en) | 2020-09-04 | 2020-09-04 | Ghost imaging optimization method based on multi-speckle pattern combination-modulation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112116675B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017030652A1 (en) * | 2015-08-20 | 2017-02-23 | Massachusetts Institute Of Technology | Sensing targets using photothermal speckle detection |
| CN106595859A (en) * | 2016-11-01 | 2017-04-26 | 清华大学 | Ghost imaging method and ghost imaging device with application of the same |
| CN110475081A (en) * | 2019-06-28 | 2019-11-19 | 中北大学 | A kind of mutative scale ghost imaging method based on transmissivity variation |
| CN110489981A (en) * | 2019-07-29 | 2019-11-22 | 西安理工大学 | Based on PCA and calculate the terrible optical image encryption method being imaged |
| CN111161234A (en) * | 2019-12-25 | 2020-05-15 | 北京航天控制仪器研究所 | Discrete cosine transform measurement basis sorting method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102015114376B4 (en) * | 2015-08-28 | 2017-10-12 | Visteon Global Technologies, Inc. | Apparatus and method for characterizing subjective speckle formation |
-
2020
- 2020-09-04 CN CN202010923096.4A patent/CN112116675B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017030652A1 (en) * | 2015-08-20 | 2017-02-23 | Massachusetts Institute Of Technology | Sensing targets using photothermal speckle detection |
| CN106595859A (en) * | 2016-11-01 | 2017-04-26 | 清华大学 | Ghost imaging method and ghost imaging device with application of the same |
| CN110475081A (en) * | 2019-06-28 | 2019-11-19 | 中北大学 | A kind of mutative scale ghost imaging method based on transmissivity variation |
| CN110489981A (en) * | 2019-07-29 | 2019-11-22 | 西安理工大学 | Based on PCA and calculate the terrible optical image encryption method being imaged |
| CN111161234A (en) * | 2019-12-25 | 2020-05-15 | 北京航天控制仪器研究所 | Discrete cosine transform measurement basis sorting method |
Non-Patent Citations (3)
| Title |
|---|
| An effective compressive computational ghost imaging with hybrid speckle pattern;Xiaoxia Wang等;《Optics Communications》;20200131;第454卷;1-7 * |
| Positive-negative corresponding normalized ghost imaging based on an adaptive threshold;LI Gao-liang等;《Laser Physics Letters》;20161231;第13卷(第11期);115202 * |
| 基于叠加散斑图的反射鬼成像;蔡宏吉等;《激光与光电子学进展》;20191231;第56卷(第7期);156-161 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112116675A (en) | 2020-12-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dreissigacker et al. | Deep-learning continuous gravitational waves: multiple detectors and realistic noise | |
| CN111127316B (en) | Single face image super-resolution method and system based on SNGAN network | |
| CN113837974B (en) | NSST domain power equipment infrared image enhancement method based on improved BEEPS filtering algorithm | |
| CN112215296B (en) | Infrared image recognition method based on transfer learning and storage medium | |
| CN112288668B (en) | Infrared and visible light image fusion method based on depth unsupervised dense convolution network | |
| Zhang et al. | VST-Net: Variance-stabilizing transformation inspired network for Poisson denoising | |
| Fei et al. | A novel visual attention method for target detection from SAR images | |
| Liu et al. | A super resolution algorithm based on attention mechanism and srgan network | |
| CN112766223A (en) | Hyperspectral image target detection method based on sample mining and background reconstruction | |
| CN113887398A (en) | GPR signal denoising method based on variational modal decomposition and singular spectrum analysis | |
| CN112116675B (en) | Ghost imaging optimization method based on multi-speckle pattern combination-modulation | |
| CN113421198B (en) | Hyperspectral image denoising method based on subspace non-local low-rank tensor decomposition | |
| Mogushi et al. | NNETFIX: an artificial neural network-based denoising engine for gravitational-wave signals | |
| CN114067217A (en) | SAR image target identification method based on non-downsampling decomposition converter | |
| Ren et al. | A global weighted least-squares optimization framework for speckle filtering of PolSAR imagery | |
| CN105718915A (en) | Face identification method and system based on multi-visual-angle typical correlation analysis | |
| CN116051444A (en) | Effective infrared and visible light image self-adaptive fusion method | |
| CN114495239A (en) | Forged image detection method and system based on frequency domain information and generation countermeasure network | |
| Roser et al. | Learning the inverse weighted radon transform | |
| Lei et al. | Improving compressed sensing image reconstruction based on atmospheric modulation using the distributed cumulative synthesis method | |
| CN117036533B (en) | Sparse angle CT reconstruction method and system for coaxial phase contrast imaging | |
| Wong | Intensity signal processing of images for optical to radar scene matching. | |
| CN115239731B (en) | Method for detecting concentration field in cylinder of internal combustion engine | |
| Tao et al. | Targets detection in SAR image used coherence analysis based on S-transform | |
| Sun et al. | An image fusion method based on region segmentation and wavelet transform |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |