CN102865889A - Detector nonlinear saturation corrective reduction method based on compressive sensing imaging system - Google Patents
Detector nonlinear saturation corrective reduction method based on compressive sensing imaging system Download PDFInfo
- Publication number
- CN102865889A CN102865889A CN2012103963488A CN201210396348A CN102865889A CN 102865889 A CN102865889 A CN 102865889A CN 2012103963488 A CN2012103963488 A CN 2012103963488A CN 201210396348 A CN201210396348 A CN 201210396348A CN 102865889 A CN102865889 A CN 102865889A
- Authority
- CN
- China
- Prior art keywords
- measured value
- detector
- saturated
- imaging system
- saturation
- 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.)
- Granted
Links
Images
Landscapes
- Image Processing (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention discloses a detector nonlinear saturation corrective reduction method based on a compressive sensing imaging system. When a detector measurement value is affected by the nonlinear saturation effect, saturation rejection correction or saturation feedback correction is selected by judging the degree of the detection value affected by the nonlinear effect. Saturation rejection correction is to remove measuring points which are greatly affected by the nonlinear effect, and saturation feedback correction is to perform feedback processing for the measuring value and correct the obtained measuring value according a given detector response curve. The processed measuring value is input into a computer for image reduction, and the obtained effect is greatly superior to the effect which is obtained through direct reduction without using the corrective reduction method. The detector nonlinear saturation corrective reduction method reduces effects of nonlinear detection error of a detector on experiment effects and improves reduction efficiency of the imaging system.
Description
Technical field
The invention belongs to technical field of imaging, particularly the saturated rectification method of reducing of a kind of detector nonlinearity based on the compressed sensing imaging system.
Background technology
After the theory of compressed sensing is suggested, the combination of compressed sensing technology and imaging system is more and more, technology also reaches its maturity, single pixel camera of inventing from people such as the Marco F.Duarte of rice university, to the 3-D imaging system based on compressed sensing, constantly have various dissimilar, the imaging system based on compressed sensing of different purposes occurs, and these systems have all demonstrated fully the benefit that the compressed sensing technology is brought, more accelerate such as image taking speed, saved complicated mechanical scanning structure etc.But when using compressed sensing that the original signal of system is reduced, at first need system to obtain one group of measured value of crossing through Gauss's matrix modulation, and the whether accurate final signal reduction effect that has directly affected system of measured value.So the accurate measurement of measured value is very crucial in based on the imaging system of a compressed sensing step.
When system detector is surveyed measured value, because the probe response of detector itself is non-linear, the measured value that can cause detecting produces non-linear saturated phenomenon, affect the accuracy of the measured value that obtains, thereby affect the reduction effect of imaging system, and as long as in detector measurement, there is nonlinear problem, if do not add processing, will produce very significantly impact to reduction effect, greatly reduce the reduction efficiency of imaging system, lower picture quality.
In traditional method of reducing based on the compressed sensing imaging system, just simply resulting measured value is reduced, such way to solve the problem is not described, if above-mentioned non-linear saturated conditions appears in detector, then image quality will be affected.So before measured value is directly inputted the method for reducing module, need to carry out pre-service according to detection event, so be subject in the situation of non-linear effects in the detector measurement value, can guarantee the also proper mass of image, improve the robustness of system.
Summary of the invention
The object of the invention is to for the problem that affects system reducing efficient based on detector nonlinearity saturation effect in the imaging system of compressed sensing, provide a kind of and reduce the detector nonlinearity response to the antidote of compressed sensing imaging system reduction effect impact, can greatly reduce non-linear saturation effect to the impact of the reduction efficiency of system by the method, improve image quality.
Realize that technical solution of the present invention is: the saturated rectification method of reducing of a kind of detector nonlinearity based on the compressed sensing imaging system, be subject in the detector measurement value under the impact of non-linear saturation effect, by judging that measured value is subjected to the degree of non-linear effects, select saturated repulsion to correct or saturated feedback correction, wherein saturated repulsion rectification is to exclude the measurement point that is subjected to non-linear effects, saturated feedback correction then is according to the explorer response curve that has provided, measured value is done feedback processing, revise measured measured value, the measured value input computing machine after processing is finished reduces and obtains going back original image.
The present invention compared with prior art, its significant advantage is: (1) by non-linear correction, has greatly reduced the detector nonlinearity effect to the impact based on the single photon image system of compressed sensing, has improved the reduction efficiency of system.(2) non-linear correction is cut from software approach, does not have complicated physical construction, has reduced the complexity of system.(3) improved the reduction rate of image.
Below in conjunction with accompanying drawing the present invention is described in further detail.
Description of drawings
Fig. 1 the present invention is based on compressed sensing imaging system device synoptic diagram.
Fig. 2 is the theory diagram that the present invention is based on detector nonlinearity response saturation feedback correction method in the photon counting imaging process of compressive sensing theory.
Fig. 3 the present invention is based on the theory diagram that detector nonlinearity response saturation in the photon counting imaging process of compressive sensing theory repels antidote.
Fig. 4 is original image before the system emulation.
Fig. 5 is that sampling rate is 30%, when not affected by detector nonlinearity, and resulting reduction picture.
Fig. 6 is that sampling rate is 30%, when affected by different nonlinear degrees, and resulting reduction picture.
Fig. 7 is that sampling rate is 30%, after not using saturated antidote (left side) and having used saturated antidote (right side), and resulting reduction picture.
Fig. 8 is that sampling rate is 30%, non-linear saturation factor is respectively 5%(figure a) in the measured measured value, 15% (figure b), 40%(schemes c), the reduction effect figure of (figure d) in the time of 80%, figure a(is left) figure b(is left) figure c(is left) figure d(is left) for not using the reduction effect of antidote, figure a(is right) figure b(is right) figure c(is right) figure d(is right) for having used the reduction effect behind the saturated repulsion antidote.
Under the different non-linear degree of saturation of Fig. 9, use saturated repulsion method (solid line) and do not use the reduction efficiency comparison diagram (the higher then key diagram of reduction efficiency picture is more near original image) of saturated repulsion method (dotted line).
Embodiment
The detector nonlinearity response rectification device that method of reducing was suitable for that the present invention is based on the compressed sensing imaging system is the imaging system device based on compressed sensing, as shown in Figure 1.Be subject in the detector measurement value under the impact of non-linear saturation effect, be subjected to the degree of non-linear effects by judging probe value, select saturated repulsion to correct or saturated feedback correction.It namely is to exclude the measurement point that is subjected to non-linear effects larger that saturated repulsion is corrected, and saturated feedback correction namely according to the explorer response curve that has provided, is done feedback processing to measured value, revises measured measured value.Measured value after will processing through the method for the invention imports the compressed sensing method of reducing
Finally gone back original image.
Method of reducing is corrected in the detector nonlinearity response that the present invention is based on the compressed sensing imaging system, and wherein the detector nonlinearity antidote may further comprise the steps:
(1) M that obtains to record from data acquisition module organizes pending measured value, and M is positive integer.
(2) response curve given according to detector judges whether measured value is in the nonlinear response zone, and establishing detector is σ '＜E＜σ for the linear probing scope of light intensity, and wherein σ ﹑ σ ' ﹑ E is light intensity; When E 〉=σ or E≤σ ', be judged as measured value and be in nonlinear state, select saturated repulsion to correct or the processing of saturated feedback correction;
(3) if the data more than or equal to 50% are subject to non-linear effects in the measured value, then use saturated feedback correction, otherwise then use saturated repulsion to correct;
(4) measurement data after will correcting imports computing machine, utilizes compression sensing method to reduce, and obtains reduction result.
Method of reducing is corrected in the detector nonlinearity response that the present invention is based on the compressed sensing imaging system, concrete processing procedure is corrected in saturated repulsion: remove the measurement matrix corresponding with the measured value of getting rid of, in the measured value of M dimension, the corresponding measurement matrix of i measured value y (i) is Φ (i, N), after i measured value is removed, measure accordingly matrix and also will be removed, i is positive integer.
Method of reducing is corrected in the detector nonlinearity response that the present invention is based on the compressed sensing imaging system, the concrete processing procedure of saturated feedback correction is: according to concrete measured value, the probe response curve that corresponding detector is given, carry out measured value compensation, obtain compensating factor, multiply each other with the measured value that records and namely to get actual offset, the response factor curvilinear function of namely supposing detector is y=ζ (y '), the response that y ' expression detector detects, y is illustrated in corresponding response factor under the measured value of y'.Then actual detector response function can be expressed as: y=ζ (y') * y', the probe value after then i reality compensates can be expressed as y (i)=ζ (y'(i)) * y'(i).
Saturated repulsion rectification and saturated feedback correction are processed as follows among the present invention:
(1) saturated feedback correction
In conjunction with Fig. 2, the response factor curvilinear function of supposing detector is y=ζ (y '), the response that y ' expression detector detects, and y is illustrated in corresponding response factor under the measured value of y '.Then actual detector response function can be expressed as: y=ζ (y') * y', the probe value after then i reality compensates can be expressed as
y(i)=ζ(y'(i))*y'(i)
Again because y'=Φ (x), so following formula can turn to y (i)=ζ (Φ x (i)) * Φ x (i) and the measured value of gained is reduced again, and can obtain to go back original image.
(2) saturated repulsion is corrected
In conjunction with Fig. 3, when detector is in the linear probing zone substantially, when only having part to be in non-linear search coverage, we use saturated repulsion method, and when detector detected luminous energy and is E, the compressed sensing measured value that we obtain can be expressed as
y＝f(E)+b
Also can be expressed as
y＝Φx+b
Y represents the response that detector detects in the formula, and b represents the trueness error of linear measurement, and Φ is the random measurement matrix, and x is input signal, and f (E) responds to the quantization function of concrete light intensity for detector.When E is in nonlinear area, very large on the impact of reconstructed results, so take the method that abandons.
If y is the measurement matrix of M*1 dimension, and the measurement number that is in the nonlinear response district in M measured value is m, and then this m measurement number will be handled as follows, and m, i, j, N are positive integer:
Y (j)=y (i) (when y (i) is in the linear response district), j
Max+ m=i
Max;
Because in the measured value of M dimension, the corresponding measurement matrix of i measured value y (i) is Φ (i, N), so after i measured value is removed, measures accordingly matrix and also will do following variation:
Φ (j, N)=Φ (i, N) (when y (i) is in the linear response district), j
Max+ m=i
Max;
Be about to original M dimension measured value y (M) and M*N dimension and measure matrix Φ (M, N), turn y(M-m into), Φ (M-m, N), the recycling method is reduced at last, obtains finally to go back original image.
Embodiment
The present invention utilizes the above-mentioned imaging system based on compressive sensing theory to realize it for the experiment of the saturated antidote of detector nonlinearity, and step is as follows:
1. open imaging system, begin to measure.
2. the M that obtains to record from data acquisition module organizes pending measured value.
3. judge that according to the measured value of type photodetector and gained pending M group measured value is subjected to the size of non-linear effects.
4. if the data greater than 50% are subject to non-linear effects in the measured value, then use saturated feedback correction.Otherwise then use saturated repulsion to correct.
5. by adopting based on l
1The protruding optimization method of Norm minimum, the measured value after processing is found the solution such as drag, and obtain final reduction result:
s.t.||y-Φψs‖＜ε
Experimental result of the present invention is by the Computer Simulation gained, and the used original image of emulation as shown in Figure 4.Fig. 7 is sampling rate 30%, reduction effect figure contrast after not using saturated feedback method (left side) and having used saturated feedback method (right side), Fig. 8 is sampling rate 30%, non-linear saturation factor is respectively 5%(figure a) in the measured value that obtains, 15% (figure b), 40%(schemes c), the reduction effect figure of (figure d) in the time of 80%, do not use the saturated repulsion method figure a(left) figure b(is left) figure c(is left) figure d(is left) and use the saturated repulsion method figure a(right) figure b(is right) figure c(is right) figure d(is right) after reduction effect figure contrast, from Fig. 7, Fig. 8 can find out and uses that the reconstruction effect of system is significantly improved behind the antidote, greatly reduced by non-linear saturated impact.
Claims (4)
1. saturated rectification method of reducing of the detector nonlinearity based on the compressed sensing imaging system, it is characterized in that being subject under the impact of non-linear saturation effect in the detector measurement value, by judging that measured value is subjected to the degree of non-linear effects, select saturated repulsion to correct or saturated feedback correction, wherein saturated repulsion rectification is to exclude the measurement point that is subjected to non-linear effects, saturated feedback correction then is according to the explorer response curve that has provided, measured value is done feedback processing, revise measured measured value, the measured value input computing machine after processing is finished reduces and obtains going back original image.
2. the saturated rectification method of reducing of the detector nonlinearity based on the compressed sensing imaging system according to claim 1 is characterized in that concrete steps are as follows:
(1) M that obtains to record from data acquisition module organizes pending measured value, and M is positive integer.
(2) response curve given according to detector judges whether measured value is in the nonlinear response zone, and establishing detector is σ '＜E＜σ for the linear probing scope of light intensity, and wherein σ ﹑ σ ' ﹑ E is light intensity; When E 〉=σ or E≤σ ', be judged as measured value and be in nonlinear state, select saturated repulsion to correct or the processing of saturated feedback correction;
(3) if the data more than or equal to 50% are subject to non-linear effects in the measured value, then use saturated feedback correction, otherwise then use saturated repulsion to correct;
(4) measurement data after will correcting imports computing machine, utilizes compression sensing method to reduce, and obtains reduction result.
3. the saturated rectification method of reducing of the detector nonlinearity based on the compressed sensing imaging system according to claim 1 and 2, it is characterized in that saturated repulsion corrects concrete processing procedure and be: remove the measurement matrix corresponding with the measured value of eliminating, in the measured value of M dimension, the corresponding measurement matrix of i measured value y (i) is Φ (i, N), after i measured value is removed, measure accordingly matrix and also will be removed, i is positive integer.
4. the saturated rectification method of reducing of the detector nonlinearity based on the compressed sensing imaging system according to claim 1 and 2, it is characterized in that the concrete processing procedure of saturated feedback correction is: according to concrete measured value, the probe response curve that corresponding detector is given, carry out measured value compensation, obtain compensating factor, multiply each other with the measured value that records and namely to get actual offset, the response factor curvilinear function of namely supposing detector is y=ζ (y'), the response that y ' expression detector detects, y is illustrated in corresponding response factor under the measured value of y'.Then actual detector response function can be expressed as: y=ζ (y') * y', the probe value after then i reality compensates is expressed as y (i)=ζ (y'(i)) * y'(i).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210396348.8A CN102865889B (en) | 2012-10-17 | 2012-10-17 | Detector nonlinear saturation corrective reduction method based on compressive sensing imaging system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210396348.8A CN102865889B (en) | 2012-10-17 | 2012-10-17 | Detector nonlinear saturation corrective reduction method based on compressive sensing imaging system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102865889A true CN102865889A (en) | 2013-01-09 |
| CN102865889B CN102865889B (en) | 2014-11-26 |
Family
ID=47444904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210396348.8A Expired - Fee Related CN102865889B (en) | 2012-10-17 | 2012-10-17 | Detector nonlinear saturation corrective reduction method based on compressive sensing imaging system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102865889B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103438993A (en) * | 2013-08-22 | 2013-12-11 | 中国科学院上海光学精密机械研究所 | Device for measuring linear interval and face response characteristics of photoelectric detector |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006122146A2 (en) * | 2005-05-10 | 2006-11-16 | William Marsh Rice University | Method and apparatus for distributed compressed sensing |
| WO2007050593A2 (en) * | 2005-10-25 | 2007-05-03 | William Marsh Rice University | Method and apparatus for signal detection, classification, and estimation from compressive measurements |
| CN101975936A (en) * | 2010-09-03 | 2011-02-16 | 杭州电子科技大学 | Rapid magnetic resonance imaging (MRI) method based on CS ( compressed sensing ) technique |
| US20110109773A1 (en) * | 2009-11-10 | 2011-05-12 | General Electric Company | System and method for adaptive nonlinear compressed visual sensing |
-
2012
- 2012-10-17 CN CN201210396348.8A patent/CN102865889B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006122146A2 (en) * | 2005-05-10 | 2006-11-16 | William Marsh Rice University | Method and apparatus for distributed compressed sensing |
| WO2007050593A2 (en) * | 2005-10-25 | 2007-05-03 | William Marsh Rice University | Method and apparatus for signal detection, classification, and estimation from compressive measurements |
| US20110109773A1 (en) * | 2009-11-10 | 2011-05-12 | General Electric Company | System and method for adaptive nonlinear compressed visual sensing |
| CN101975936A (en) * | 2010-09-03 | 2011-02-16 | 杭州电子科技大学 | Rapid magnetic resonance imaging (MRI) method based on CS ( compressed sensing ) technique |
Non-Patent Citations (3)
| Title |
|---|
| E. CANDES 等: "Stable signal recovery from incomplete", 《COMMUN. PURE AND APPL. MATH》, vol. 59, no. 8, 31 December 2006 (2006-12-31), pages 1207 - 1223 * |
| H. T. KUNG 等: "Identifying Bad Measurements in Compressive Sensing", 《PROCEEDINGS OF IEEE CONFERENCE ON COMPUTER COMMUNICATIONS WORKSHOPS (INFOCOM WKSHPS)》, 31 December 2011 (2011-12-31) * |
| 陆利忠 等: "《现代电子线路基础（下册）》", 31 January 2011, article "现代电子线路基础（下册）" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103438993A (en) * | 2013-08-22 | 2013-12-11 | 中国科学院上海光学精密机械研究所 | Device for measuring linear interval and face response characteristics of photoelectric detector |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102865889B (en) | 2014-11-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103108108B (en) | Image stabilizing method and image stabilizing device | |
| US10488355B2 (en) | Gas concentration-thickness product measurement device, gas concentration-thickness product measurement method, and computer-readable recording medium having gas concentration-thickness product measurement program recorded thereon | |
| CN104580894B (en) | The multi-point correcting method and system of a kind of infrared focus plane | |
| CN103335724B (en) | Calibration-based scene self-adaption IRFPA heterogeneity correction method | |
| JPWO2018198702A1 (en) | Stress measuring device, stress measuring system and stress measuring method | |
| CN104067118B (en) | A kind of temperature sensor numerical compensation method, device and air quality detector | |
| CN104468916A (en) | Test method based on cell phone response time | |
| CN103226688B (en) | The authentication method of the anti-tamper and anti-counterfeiting of a kind of Quick Response Code | |
| CN102509304A (en) | Intelligent optimization-based camera calibration method | |
| CN104008543A (en) | Image fusion quality evaluation method | |
| CN104515599A (en) | FPGA-based infrared focal plane array blind pixel detection system and FPGA-based infrared focal plane array blind pixel detection method | |
| Sheng et al. | Calibration-based NUC method in real-time based on IRFPA | |
| CN102692273B (en) | Method of on-track detection for MTF (modulation transfer function) of interference hyperspectral imager | |
| CN104462778A (en) | PM2.5 pollutant measurement method based on deep learning | |
| CN103941605A (en) | Sensor synchronous controller for bridge monitoring | |
| CN103837828A (en) | Travel curve test method and system of camera-shooting mode motor and mobile terminal | |
| PL435081A1 (en) | Radiation image correction method, correction apparatus and correction system | |
| Zheng et al. | Atmospheric correction on CCD data of HJ-1 satellite and analysis of its effect | |
| CN104113674B (en) | Image real-time correction output method for linear array CCD imaging system based on FPGA | |
| CN102865889B (en) | Detector nonlinear saturation corrective reduction method based on compressive sensing imaging system | |
| CN106482648A (en) | Based on the absolute monitoring device of thin tail sheep in the long-distance plane of fixed point and method | |
| CN204594468U (en) | Aircraft flight error correction device and unmanned vehicle | |
| CN102410844A (en) | Method and device for correcting non-uniformity of image of high-dynamic star sensor | |
| CN104748865A (en) | Multipoint combined correction method used for infrared image | |
| CN104376531A (en) | Image splicing method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20141126 Termination date: 20181017 |