CN102768069B - Single-photon spectral counting and imaging system and method for complementary measurement - Google Patents

Abstract

The invention provides a single-photon spectral counting and imaging system and method for complementary measurement. According to the system, a plurality of single-photon dot detection units corresponding to different wavelengths are respectively selected in a visible range and a near infrared range so as to form two linear arrays which are distributed in exit directions of two arms of a spatial light modulator together with a spectral light splitter. The method comprises the following steps that: the spatial light modulator reflects the light imaged on the spatial light modulator to the two arms and respectively collimates and splits the light, visible spectrum is taken by one arm, and near infrared spectrum is taken by the other arm, so that the light can be detected by the corresponding linear array of the single-photon detector. According to the wavelength information corresponding to a complementary matrix, count values and a channel, visible and near infrared color images can be rebuilt with a related compressed sensing algorithm and a spectrogram for wavelength-optical power can be counted for spectral analysis. The system and the method disclosed by the invention solve the problem that the existing imaging spectrograph cannot simultaneously detect spectra of the visible light and the near infrared light, and have the advantages of low dimension, high flux, high sensitivity, high resolution and the like.

Description

A kind of single photon spectrum counting imaging system and method for complementary measurement

Technical field

The present invention relates to the technical field of utmost point low light level imaging spectral, particularly a kind of single photon spectrum of complementary measurement counting imaging system and method.

Background technology

In emerging high-tech areas such as light spectrum image-forming Measurement and analysis, biological detection, starry sky detection, night vision observation, utmost point weak light detection, concentration detection, remotely sensed image, imaging of medical, monitoring, not only need to survey the utmost point low light level, also to carry out component distributing analysis to several spectral bands, and the imaging spectral of visible ray is surveyed and cannot be satisfied the demands, and scientists more often needs to know the spectral information of near infrared light wave band.In addition, existing imaging spectral instrument is insensitive to the utmost point low light level, mostly adopts scan mode, and poor stability is difficult to reach the requirement of imaging high precision.Because laser instrument is generally operational on the wavelength of frequency multiplication, wavelength coverage is limited, conventional imaging detector is also only operated in visible light wave range, be difficult to realize when utmost point low light level object is carried out to the wide range component distributing of visible ray and near infrared light and detect and analyze, urgently a kind of brand-new imaging spectrum system meets growing scientific research demand.

The present invention arises at the historic moment for addressing the above problem, and is also improvement and the innovation based on previous work simultaneously.In this area, this research institute has a Patents: " a kind of utmost point low light level multispectral imaging method and system thereof " (application number or patent No.: 201110166471.6, applicant or patentee: Space Sci. & Application Research Center, Chinese Academy of Sciences), so in this patented claim due to do not consider the feature of random measurement matrix only adopt an arm reflected light path of spatial light modulator (SLM) carry out photon collect light splitting, and the algorithm reconstruction quality adopting is poor, operand is large, in principle, remain in certain technical defect, the present invention takes into full account that in order to overcome above-mentioned technological deficiency the relation between technical characterictic proposes a kind of single photon spectrum counting imaging system and method for complementary measurement, adopt both arms complementary measurement, visible ray and near infrared light are measured simultaneously, principle and algorithm are improved, and the spectrum map generalization method of wavelength-luminous power proposed, to solve above-mentioned a series of problem.

Light spectrum image-forming is obtain and show the important technology of accurate colouring information, and one of reason is that spectrum picture has comprised spectral information, and former two is therefore that light spectrum image-forming technology has overcome metamerism phenomenon well.Light spectrum image-forming to utmost point low light level object has wide practical use at multiple fields especially.

Photon counting imaging is exactly a kind of utmost point weak light detection technology, and by being recorded as the photon counting of image position, counting is equivalent to intensity signal, so obtain the space distribution situation of light intensity, finally adds up piece image.

Existing imaging spectral technology must be by any one dimension in one dimension spectrum, two-dimensional image, mode with sub-scanning realizes, due to poor stability, the scanning process complexity of nanometer displacement scanning platform, not only increase manufacturing cost, also the test duration that has greatly extended object under test, for some biological sample, can be lowered into power.The drawback of bringing is like this sampling time must reserve sufficient space to scanning.If when one dimension spectrum is realized with scanning form, must adopt bin detector (being formed by a large amount of photosensitive detecting elements) to survey, as ICCD, EMCCD, APDs etc.The bin detector certain hour (integral time) that need to expose in the time that the utmost point low light level is surveyed, average luminous flux to unit picture element is minimum, adds the impact of dark counting, and signal to noise ratio (S/N ratio) is poor, extremely difficult accurately reckoning drops on the light intensity value on this pixel, has the problem of sensitivity.Wherein ICCD, EMCCD are known as and can accomplish single photon detection, but need degree of depth semiconductor refrigerating, cost costliness, ICCD spatial resolution is poor, and EMCCD spatial resolution is slightly good, the common issue of existence is all under the low light level, to be difficult to noise of instrument to control or linear output; And APDs can be operated in Geiger mode angular position digitizer, but be still in conceptual phase, and high-precision APD array is difficult to manufacture and flow, the APD array that Lincoln laboratory is externally announced also only has 64 × 256 pixels, and to China's embargo, and current APD array wavelength response range is very limited, only in visual wave band reach quantum efficiency peak value, because each pixel is minimum, the necessary mean allocation of luminous flux is on whole array, and the impact of shot noise will become very remarkable so.

Described compressed sensing (Compressive Sensing, being called for short CS) theory is the brand-new mathematical theory being proposed by the people such as Donoho, Candes and Tao, realize with the mode of stochastic sampling, data sampling number (far below the limit of Nyquist/Shannon's sampling theorem) still less and carry out ideally restoring signal, and there is more high robust.The first compression sampling of this principle, is shone upon measured signal it is sampled to low-dimensional by higher-dimension, chooses suitable sparse transformation framework Ψ, makes x through Ψ conversion gained vector be sparse, then, according to observation data y, measurement matrix A and framework Ψ, solve finally by be finally inversed by x.

Described spatial light modulator (Spatial Light Modulator, be called for short SLM) be that a class can load on information on the optical data field of one dimension or bidimensional, under the control of time dependent electric drive signal or other signal, the Real-Time Optical information processing device that can modulate photodistributed intensity on space, the light of the angle incident specifying by spatial light modulator can be reflexed to both direction with certain probability, the general state that adopts two-value to characterize reflection direction, common spatial light modulator has Digital Micromirror Device, liquid crystal light valve, frosted glass etc.Wherein, Digital Micromirror Device (Digital Micro-mirror Device is called for short DMD), is the most accurate in the world photoswitch.The core of DMD is arranged on the micro mirror array that the micro-reflector on hinge form (DMD of main flow is by 1024 × 768 array formation by thousands of, maximum can be to 2048 × 1152), each eyeglass is of a size of 14 μ m × 14 μ m(or 16 μ m × 16 μ m) and light that can a pixel of break-make, these micro mirrors are all suspending, carry out electronic addressing by the storage unit under each eyeglass with scale-of-two planed signal, just can allow each eyeglass tilt to both sides with electrostatic means, this two states is designated as to 1 and 0, respectively corresponding " opening " and " pass ", in the time that eyeglass is not worked, they are in " berthing " state of 0 °.

How to realize the high-quality image-forming spectral measurement to utmost point low light level object, how to evade the restriction of mechanical scanning to imaging resolution, how to reduce dimension and the measurement scale measured, this proposes new challenge to imaging spectral, the present invention makes up the defect of prior art in signal to noise ratio (S/N ratio) with high flux, survey the observation that realizes visible ray and near infrared light with both arms, use compressed sensing theory to improve image quality and reduce measurement scale.

Summary of the invention

The object of the invention is to, for realizing, the spectral component of observed objects visible ray and near infrared light is carried out to evaluating objects simultaneously, solve dimension scanning, reconstruction precision that routine spectra formation method exists poor, to problems such as the utmost point low light level are insensitive, thereby a kind of single photon spectrum counting imaging system and method for complementary measurement are provided.

The present invention is taking compressed sensing theory as basis, using visible ray single-photon detector linear array and near infrared light single-photon detector linear array as detecting element, adopt complementary measurement technology and spectrum light splitting technology by reflected light collection simultaneously, light splitting, the detection of spatial light modulator two arms, then utilize compressed sensing thought to rebuild, recover the coloured image of visible ray and near infrared light, the line correlation spectral analysis of going forward side by side.

For achieving the above object, the invention provides a kind of single photon spectrum counting imaging system of complementary measurement, it is characterized in that, described system comprises:

Optical imagery parts, spatial light modulator, first group of receipts optical alignment parts, second group of receipts optical alignment parts, first group of spectrum light splitting part, second group of spectrum light splitting part, visible ray single-photon detector linear array, near infrared light single-photon detector linear array, hyperchannel counter and information process unit; Wherein, the light that carries object under test information is imaged in described spatial light modulator by described optical imagery parts, and the emergent light after the random optical modulation of described spatial light modulator is divided into two arm direction transmission;

Receive optical alignment parts, first group of spectrum light splitting part and visible ray single-photon detector linear array and be arranged at the first arm direction, for surveying the photon of visible light, export some group pulse waveforms for described first group; Receive optical alignment parts, second group of spectrum light splitting part and near infrared light single-photon detector linear array and be arranged at the second arm direction, for surveying the photon of near infrared light spectrum, export some group pulse waveforms for described second group;

Described hyperchannel counter is arranged at via on the circuit after described visible ray single-photon detector linear array and near infrared light single-photon detector linear array, carries out filtering, discriminating and stored count for the pulse waveform under each passage;

Described information process unit is according to the corresponding wavelength information of each counting channel on the each count value on the complementary matrix in described spatial light modulator, hyperchannel counter and hyperchannel counter, use associated compressed sensing algorithm to reconstruct the coloured image of visible ray and near infrared light, and can be further analyzed the composition under specific wavelength.

In technique scheme, described visible ray single-photon detector linear array and near infrared light single-photon detector linear array are operated in by several corresponding different wave lengths the linear array that the single photon point probe under Geiger mode angular position digitizer forms, select respectively the single photon point detecting element of several corresponding different wave lengths for visible light wave range and near infrared light wave band, many group optical filters of can arranging in pairs or groups use;

Described visible ray single-photon detector linear array and near infrared light single-photon detector linear array position can phase double replacements;

The detection mode of described visible ray single-photon detector linear array and near infrared light single-photon detector linear array adopts fiber optic collimator or Free-space coupling;

Wherein, visible ray single-photon detector linear array and near infrared light single-photon detector linear array also can enable certain a line or a certain row obtain by avalanche photodiode arrays.

In technique scheme, when the every conversion of the stochastic matrix in spatial light modulator once, described hyperchannel counter records is once counted, the measured value in the compressed sensing algorithm that some batch total numerical value of its output adopt as information process unit.

In technique scheme, described first group of spectrum light splitting part and second group of spectrum light splitting part include: optical alignment part, light splitting part, measurement of angle part, photometric observation and measure portion; Light absorbing part part and described optical alignment part are used in conjunction with, for the emergent light in a certain arm direction in spatial light modulator is all collected and preliminary collimation.

Optimize, in the time that utmost point low light level light intensity exceeds the investigative range of described visible ray single-photon detector linear array or near infrared light single-photon detector linear array, on any light path position before described visible ray single-photon detector linear array or near infrared light single-photon detector linear array, attenuator is set, for light decay being reduced to the investigative range of described single-photon detector linear array.

The single photon spectrum counting formation method that a kind of complementary measurement is also provided based on said system the present invention, described method comprises:

Step 101) light that carries object under test information is imaged in spatial light modulator by optical imagery parts, and through the random optical modulation of described spatial light modulator, emergent light is divided into two arm directions with certain probability;

Step 102) respectively light beam is carried out light splitting and assembles and collect in two arm exit directions, be collected in visible ray single-photon detector linear array and near infrared light single-photon detector linear array, be recorded in hyperchannel counter with photon counting form;

Step 103) coloured image of realizing visible ray and near infrared light by compressed sensing algorithm rebuilds or correlation spectrum analysis, completes the single photon spectrum counting imaging of complementary measurement.

In technique scheme, described step 103) further comprise:

Step 103-1), the step for compression sampling:

By spatial light modulator, the light being incident on it is carried out to random optical modulation, make two-way emergent light invest follow-up receipts optical alignment parts and single-photon detector linear array with certain probability respectively, the every conversion of random array in spatial light modulator once, measure simultaneously and record the corresponding total photon number of some specific wavelengths on two bands, be equivalent to the light intensity of some points on measure spectrum band, measure M time, M is much smaller than measured signal dimension N, to complete measured signal compression sampling in visible light wave range and near infrared light wave band;

Wherein, described spatial light modulator refers under the control of time dependent signal, the Real-Time Optical information processing device that can modulate photodistributed intensity on space, the light of the angle incident specifying by spatial light modulator can be reflexed to both direction with certain probability, what described random optical modulation referred to modulation matrix in spatial light modulator is very random, and this matrix adopts two-value to characterize the state of reflection direction;

Step 103-2), the step for sparse reconstruction:

Total photon number that visible ray single-photon detector linear array and near infrared light single-photon detector linear array are detected in twice transfer interval of spatial light modulator random array is as two groups of measured value y _iand y ' _i, corresponding stochastic matrix a respectively _ibenefit a ' with matrix _i, due to 0 and 1 two value representations for matrix element, the benefit of matrix is I-a, wherein I is unit matrix;

By a _iand a ' _ibe stretched to respectively a line, respectively as visible ray measure matrix A and near infrared light measure matrix A ' in i capable, measure M time, measure matrix all to amount to M capable for two, if visible ray single-photon detector linear array has P point, near infrared light single-photon detector linear array has Q point, the measured value of two arms has respectively P group and Q group, every group all amounts to M dimension, if two arms are respectively specifically chosen some groups of data, be more than or equal to three groups, utilize associated compressed sensing reconstruction algorithm to rebuild the coloured image of object of observation at visible light wave range and near infrared light wave band, wherein the coloured image of near infrared light wave band is marked, choose from big to small three kinds of infrared wavelengths by wavelength and be designated red green blue tricolor, get one group of data if appoint and just can carry out the constituent analysis of this wavelength.

In technique scheme, the counting of each passage is added up, photon counting is converted into luminous power, in conjunction with the corresponding wavelength information of each passage, just can count the spectrogram of a wavelength-luminous power, and then obtain the characteristic peak of observed objects.

In technique scheme, the mathematical model of described compressed sensing is as follows:

Get a specific wavelength at visible-range and each of near infrared range, the signal under this two wavelength is stretched to row: sampling process can be regarded two groups of processes of measuring matrixes and two specific wavelength signals and doing inner product as, obtains respectively two groups of observation vector y _pi, y ' _qi, i=1,2 ..., M, p=1,2 ..., P, q=1,2 ..., Q, wherein (K<M < < N), system noise, if x be compressible or can sparse expression, wherein Ψ=[ψ ₁, ψ ₂..., ψ _n] be sparse transformation matrix, be associated degree of rarefication, so, the process of compression sampling can be described as following formula:

$[y_p,y_q^{\&prime;}]=[\mathrm{A\&Psi;}{\stackrel{\&OverBar;}{x}}_p+e,A^{\&prime;}\mathrm{\&Psi;}{\stackrel{\&OverBar;}{x}}_q^{\&prime;}+e^{\&prime;}]$

Wherein, A is set to two-value random measurement matrix;

Described sparse reconstruction is at known observation data y _p, y ' _qwith measurement matrix A, under the condition of A ', solve x _p, x ' _q, solving with the strategy of compressed sensing, specific algorithm is described as following formula:

$\underset{{\stackrel{\&OverBar;}{x}}_p}{\min }\frac{1}{2}{\left|\right|y_p-\mathrm{A\&Psi;}{\stackrel{\&OverBar;}{x}}_p\left|\right|}_2^2+\mathrm{\&tau;}{\left|\right|{\stackrel{\&OverBar;}{x}}_p\left|\right|}_1,\underset{{\stackrel{\&OverBar;}{x}}_q^{\&prime;}}{\min }\frac{1}{2}{\left|\right|y_q^{\&prime;}-A^{\&prime;}\mathrm{\&Psi;}{\stackrel{\&OverBar;}{x}}_q^{\&prime;}\left|\right|}_2^2+\mathrm{\&tau;}{\left|\right|{\stackrel{\&OverBar;}{x}}_q^{\&prime;}\left|\right|}_1$

Wherein, || ... || _lrepresent norm operator, only need the inferior measurement of M≤O (Klog (N/K)), just can be finally inversed by the light intensity space distribution information x of object under test under visible light wave range and two specific wavelengths of near infrared light wave band _pand x ' _qif, specifically choose separately 3 groups of measurement data at visible ray and near infrared light wave band, reuse above-mentioned model, just adopt three primary color theory can reconstruct the coloured image of visible light wave range and near infrared light wave band.

In technique scheme, described associated compressed sensing algorithm is based on two measurement matrixes that meet complementary relationship, core algorithm strategy is regarded unit matrix to measure matrix participation computing as, described core algorithm is based on compressed sensing algorithm, after computing, add corrected parameter, the income value after correcting process again with x _pdo match operation, just obtain x ' _q, make visible light signal x _pwith near infrared light signal x ' _qcan obtain simultaneously.

Compared with prior art the invention has the advantages that: the present invention is with compressed sensing theory, complementary measurement technology and spectrum light splitting technology are basis, by the random optical modulation of spatial light modulator, using visible ray single-photon detector linear array and near infrared light single-photon detector linear array as detecting element, carry out both arms detection, realize with linear array sampling dimension and obtained space two-dimensional, the three-dimensional information parameters such as spectrum dimension, greatly save and survey dimension, can measure utmost point low light level visible ray information and near infrared light information simultaneously, can obtain more spectral information, can carry out object constituent analysis to any interested wavelength, contribute to advance the fast development of association area, sensitivity can reach single photon level, resolution is high.Adopt the mode of complementary measurement, can dwindle calculated amount, the feature that Matrix Complementarity is measured in improved associated compressed sensing algorithm utilization improves algorithm reconstruction precision, and can the larger image of calculating pixel.The present invention has expanded near infrared light investigative range on existing basis, and visible ray and near infrared light are surveyed and can be carried out simultaneously, in special forward position, scientific research field has broad prospect of application, without scanning, directly linear array imaging, machinery-free vibration effect, luminous flux is high, signal to noise ratio (S/N ratio) is high, and wavelength wide coverage has solved the contradiction between the supply and demand of spectrum field.

Based on above advantage and innovative point, the present invention can be widely used in the fields such as light spectrum image-forming Measurement and analysis, biotic component detection, astronomical sight, night vision observation, utmost point weak light detection, concentration detection, imaging of medical, military and national defense, monitoring, remotely sensed image, terrible imaging.

Brief description of the drawings

Fig. 1 is the structural representation of the single photon spectrum counting imaging system of complementary measurement of the present invention;

Fig. 2 is the simulated experiment result of the embodiment of the present invention, and M17SWex nebula visible images red green blue tricolor decomposes and image reconstruction; Wherein, (a) be the R of M17SWex nebula visible ray original image divide spirogram, (b) be the G of M17SWex nebula visible ray original image to divide spirogram, (c) be that the B of M17SWex nebula visible ray original image divides spirogram, be (d) colour original of M17SWex nebula visible ray original image; (e) being that reconstruction figure, (f) of visible ray R component is that reconstruction figure, (g) of visible ray G component is the reconstruction figure of visible ray B component, is (h) the visible ray cromogram being finally inversed by according to RGB three-component;

Fig. 3 is the simulated experiment result of the embodiment of the present invention, and M17SWex nebula infrared light image red green blue tricolor decomposes and image reconstruction; Wherein, (a) and (b), (c), (d) are respectively R, G, B component and the colour originals of M17SWex nebula infrared light original image; (e), (f), (g) be respectively the reconstruction figure of infrared light R, G, B component, is (h) the infrared light cromogram being finally inversed by according to this three-component.

Accompanying drawing mark

1, optical imagery parts 2, spatial light modulator

3, receive optical alignment parts 4, second group and receive optical alignment parts for first group

5, first group of spectrum light splitting part 6, second group of spectrum light splitting part

7, visible ray single-photon detector linear array 8, near infrared light single-photon detector linear array

9, hyperchannel counter

Embodiment

Below in conjunction with accompanying drawing, the present invention is described in further detail.

The single photon spectrum counting imaging system of a kind of complementary measurement that the present invention proposes, it is characterized in that, the single photon spectrum counting imaging system of this complementary measurement adopts compressed sensing theory, complementary measurement technology and spectrum light splitting technology, and using visible ray single-photon detector linear array and near infrared light single-photon detector linear array as detecting element, the multispectral picture that is counted as of single photon, the coloured image of output visible ray and near infrared light, and can carry out correlation spectrum analysis;

The single photon spectrum counting imaging system of described complementary measurement comprises: optical imagery parts, spatial light modulator, two receipts optical alignment parts, two spectrum light splitting parts, visible ray single-photon detector linear array, near infrared light single-photon detector linear array and hyperchannel counters;

Wherein, the light that carries object under test information is imaged in spatial light modulator (SLM) by optical imagery parts, through the random optical modulation of spatial light modulator, emergent light is divided into two arm directions with certain probability, first respectively the light on two arms is collimated and spectrum light splitting, one arm is got visible light, one arm is got near infrared light spectrum, collect respectively into visible ray single-photon detector linear array and near infrared light single-photon detector linear array, and be recorded in several passages in hyperchannel counter with counting form, last according to the complementary matrix in spatial light modulator, the corresponding wavelength information of many batch totals numerical value and passage on hyperchannel counter, use associated compressed sensing algorithm can reconstruct the coloured image of visible ray and near infrared light, and can be further analyzed the composition under specific wavelength.

Described single-photon detector linear array is operated in by several corresponding different wave lengths the linear array that the single photon point probe under Geiger mode angular position digitizer forms, possesses single photon resolution characteristic, select respectively the single photon point detecting element of several corresponding different wave lengths at visible light wave range and near infrared light wave band, many group optical filters of can arranging in pairs or groups use, in linear array of composition of visible-range, another linear array of the composition of near infrared range, be separately positioned on two arm exit directions of spatial light modulator, both positions can phase double replacement, detection mode can adopt fiber optic collimator or Free-space coupling, described single-photon detector linear array also can enable certain a line or the acquisition of a certain row by avalanche diode (APD) array.

Described single-photon detector linear array need coordinate hyperchannel counter to use, single-photon detector linear array is for detection of photons, export some group pulse waveforms, hyperchannel counter is for carrying out filtering, discriminating and stored count to the pulse waveform under each passage, the every conversion of stochastic matrix in spatial light modulator once, record is counting once, and some batch total numerical value of its output are as the measured value in compressed sensing algorithm.

Described spectrum light splitting part comprises: optical alignment part, light splitting part, measurement of angle part, photometric observation and measure portion, for spectrum light splitting and Measurement and analysis, generally adopt prismatic decomposition or grating beam splitting; Receive optical alignment parts and coordinate it to use, for the emergent light in a certain arm direction in spatial light modulator is all collected and preliminary collimation.

Optionally, in the time that utmost point low light level light intensity exceeds the investigative range of single-photon detector linear array, need on any light path position before single-photon detector linear array, suitable attenuator be set, for light decay being reduced to the investigative range of single-photon detector linear array, if utmost point low light level light intensity is in the investigative range of single-photon detector linear array, without attenuator is set again.

The present invention also proposes a kind of single photon spectrum counting formation method of complementary measurement, it is characterized in that, the method has adopted compressed sensing theory, complementary measurement technology and spectrum light splitting technology, respectively light beam is carried out light splitting and assembles and collect in spatial light modulator two arm exit directions, be collected in visible ray single-photon detector linear array and near infrared light single-photon detector linear array, be recorded in hyperchannel counter with photon counting form, realize the coloured image of visible ray and near infrared light rebuilds and correlation spectrum analysis by compressed sensing algorithm, described method comprises following steps:

Step 1, the step for compression sampling:

By spatial light modulator, the light being incident on it is carried out to random optical modulation, make two-way emergent light invest follow-up receipts optical alignment parts and single-photon detector linear array with certain probability respectively, the every conversion of random array in spatial light modulator once, measure simultaneously and record the corresponding total photon number of some specific wavelengths on two bands, be equivalent to the light intensity of some points on measure spectrum band, measure M time, M is much smaller than measured signal dimension N, to complete measured signal compression sampling in visible light wave range and near infrared light wave band;

Described spatial light modulator refers under the control of time dependent electric drive signal or other signal, the Real-Time Optical information processing device that can modulate photodistributed intensity on space, the light of the angle incident specifying by spatial light modulator can be reflexed to both direction with certain probability, its direction also needs not be certain fixed angle, can be limited within the scope of certain angle, what described random optical modulation referred to modulation matrix in spatial light modulator is very random, the general state that adopts two-value to characterize reflection direction, common spatial light modulator has Digital Micromirror Device (DMD), liquid crystal light valve, frosted glass etc.,

Step 2, the step for sparse reconstruction:

Total photon number that visible ray single-photon detector linear array and near infrared light single-photon detector linear array are detected in twice transfer interval of spatial light modulator random array is as two groups of measured value y _iand y ' _i, corresponding stochastic matrix a respectively _ibenefit a ' with matrix _i, due to 0 and 1 two value representations for matrix element, the benefit of matrix is I-a, wherein I is unit matrix, by a _iand a ' _ibe stretched to respectively a line, respectively as visible ray measure matrix A and near infrared light measure matrix A ' in i capable, measure M time, measure matrix all to amount to M capable for two, if visible ray single-photon detector linear array has P point, near infrared light single-photon detector linear array has Q point, the measured value of two arms has respectively P group and Q group, every group all amounts to M dimension, if two arms are respectively specifically chosen 3 groups of data, utilize associated compressed sensing reconstruction algorithm just can reconstruct the coloured image of object of observation at visible light wave range and near infrared light wave band, wherein the coloured image of near infrared light wave band is marked, choose from big to small three kinds of infrared wavelengths by wavelength and be designated red green blue tricolor, get one group of data if appoint and just can carry out the constituent analysis of this wavelength.

The counting of each passage is added up, photon counting is converted into luminous power, in conjunction with the corresponding wavelength information of each passage, just can count the spectrogram of a wavelength-luminous power, and then obtain the characteristic peak of observed objects.

Its mathematical model is as follows:

Get a specific wavelength at visible-range and each of near infrared range, the signal under this two wavelength is stretched to row: sampling process can be regarded two groups of processes of measuring matrixes and two specific wavelength signals and doing inner product as, obtains respectively two groups of observation vector y _pi, y ' _qi, i=1,2 ..., M, p=1,2 ..., P, q=1,2 ..., Q, wherein (K<M < < N), system noise, if x be compressible or can sparse expression, wherein Ψ=[ψ ₁, ψ ₂..., ψ _n] be sparse transformation matrix (being sparse framework), be associated degree of rarefication, so, the process of compression sampling can be described as following formula:

$[y_p,y_q^{\&prime;}]=[\mathrm{A\&Psi;}{\stackrel{\&OverBar;}{x}}_p+e,A^{\&prime;}\mathrm{\&Psi;}{\stackrel{\&OverBar;}{x}}_q^{\&prime;}+e^{\&prime;}]$

Wherein A Ψ, A ' Ψ all needs to meet Restricted Isometry Property(RIP), A, A ' and Ψ need be uncorrelated, and due to A, A ' complementation, for meeting above-mentioned condition, only needs A to be set to two-value random measurement matrix in the present invention;

Described sparse reconstruction is at known observation data y _p, y ' _qwith measurement matrix A, under the condition of A ', solve x _p, x ' _q, this is a NP-hard problem, but is converted into l1 norm or l2 norm problem, just can solve by the thought of compressed sensing, algorithm can divide numerous species type, as example, gets wherein a kind of common form of presentation, is described as following formula:

$\underset{{\stackrel{\&OverBar;}{x}}_p}{\min }\frac{1}{2}{\left|\right|y_p-\mathrm{A\&Psi;}{\stackrel{\&OverBar;}{x}}_p\left|\right|}_2^2+\mathrm{\&tau;}{\left|\right|{\stackrel{\&OverBar;}{x}}_p\left|\right|}_1,\underset{{\stackrel{\&OverBar;}{x}}_q^{\&prime;}}{\min }\frac{1}{2}{\left|\right|y_q^{\&prime;}-A^{\&prime;}\mathrm{\&Psi;}{\stackrel{\&OverBar;}{x}}_q^{\&prime;}\left|\right|}_2^2+\mathrm{\&tau;}{\left|\right|{\stackrel{\&OverBar;}{x}}_q^{\&prime;}\left|\right|}_1$

Wherein || ... || _lrepresent norm operator, only need the inferior measurement of M≤O (Klog (N/K)), just can be finally inversed by the light intensity space distribution information x of object under test under visible light wave range and two specific wavelengths of near infrared light wave band _pand x ' _qif, specifically choose separately 3 groups of measurement data at visible ray and near infrared light wave band, reuse above-mentioned model, just adopt three primary color theory can reconstruct the coloured image of visible light wave range and near infrared light wave band.

Optionally, described compressed sensing algorithm comprises: greedy reconstruction algorithm, coupling track algorithm MP, orthogonal coupling track algorithm OMP, base track algorithm BP, LASSO, LARS, GPSR, Bayesian Estimation algorithm, magic, IST, TV, StOMP, CoSaMP, LBI, SP, l1_ls, smp algorithm, SpaRSA algorithm, TwIST algorithm, l ₀reconstruction algorithm, l ₁reconstruction algorithm, l ₂reconstruction algorithm etc.; The core concept of algorithm communicates, and is alternating minimization and upgrades multiplier, in the skill of calculating, can slightly have any different, and does the explanation of algorithm core concept as an example of TV algorithm example:

TV model is: s.t.Au=b, is equivalent to s.t.Au=b and D _iu=w _i; Corresponding augmentation lagrange problem is:

$\underset{w_i,u}{\min }\underset{i}{\mathrm{\&Sigma;}}({\left|\right|w_i\left|\right|}_2-v_i^T(D_{i}u-w_i)+\frac{\mathrm{\&beta;}}{2}{\left|\right|D_{i}u-w_i\left|\right|}_2^2)-{\mathrm{\&lambda;}}^T(\mathrm{Au}-b)+\frac{\mathrm{\&mu;}}{2}{\left|\right|\mathrm{Au}-b\left|\right|}_2^2$

Input b, A, μ, β >0, initialization u=b;

In the time not restraining, be similar to and minimize augmentation Lagrange power function by alternating direction scheme (alternating direction scheme), and constantly update multiplier, repeat following three steps operations:

Fixing u is constant for step (1), is calculated as follows w

$w_i=\max \{{\left|\right|D_{i}u-v_i/\mathrm{\&beta;}\left|\right|}_2-\frac{1}{\mathrm{\&beta;}},0\}\frac{D_{i}u-v_i/\mathrm{\&beta;}}{{\left|\right|D_{i}u-v_i/\mathrm{\&beta;}\left|\right|}_2}$

Fixing w is constant for step (2), is calculated as follows u

$u=F^{-1}\left(\frac{F\left(D^{\left(1\right)}\right)*\mathrm{oF}\left(w_1\right)+F{\left(D^{\left(2\right)}\right)}^{*}\mathrm{oF}\left(w_2\right)+(\mathrm{\&mu;}/\mathrm{\&beta;})F{\left(A\right)}^{*}\mathrm{oF}\left(b\right)}{F\left(D^{\left(1\right)}\right)*\mathrm{oF}\left(D^{\left(1\right)}\right)+F{\left(D^{\left(2\right)}\right)}^{*}\mathrm{oF}\left(D^{\left(2\right)}\right)+(\mathrm{\&mu;}/\mathrm{\&beta;})F{\left(A\right)}^{*}\mathrm{oF}\left(A\right)}\right)$

Wherein, D ⁽¹⁾and D ⁽²⁾be respectively single order finite matrix level difference and vertical differentiation, F is two dimensional discrete Fourier transform, ^*represent complex conjugate, o represents that number takes advantage of;

Step (3) is upgraded multiplier: $v_i\&LeftArrow;v_i-\mathrm{\&beta;}(D_i\hat{u}-{\hat{w}}_i),$ $\mathrm{\&lambda;}\&LeftArrow;\mathrm{\&lambda;}-\mathrm{\&mu;}(A\hat{u}-b).$

Described associated compressed sensing algorithm is that to measure matrixes based on two be complementary and propose, and core algorithm thought is unit matrix to regard as and measure matrix and participate in computing, adds corrected parameter on the basis of described algorithm idea, income value again with x _pdo match operation, just obtain x ' _q, very big easy computing, dwindles storage space, visible light signal x _pwith near infrared light signal x ' _qcan obtain simultaneously.

As shown in Figure 1, the utmost point low light level that carries observed objects information is imaged in spatial light modulator 2 through optical imagery parts 1.Spatial light modulator can carry out random optical modulation to incident light, make it with certain probability outgoing to two arm direction, in this two arms exit direction, arrange respectively first group and receive optical alignment parts 3 and second group of receipts optical alignment parts 4, first group of spectrum light splitting part 5, second group of spectrum light splitting part 6, visible ray single-photon detector linear array 7 and near infrared light single-photon detector linear array 8, receive optical alignment parts for the reflected light of two arms being collected and preliminary collimation for two groups, two groups of spectrum light splitting parts further collimate the light of two arms, and then point spectral measurement, be incident upon in single-photon detector linear array.It should be noted that, the position of visible ray single-photon detector linear array 7 and near infrared light single-photon detector linear array 8 is replaceable.The pulse waveform of single-photon detector linear array output is input in some counting channels of hyperchannel counter 9, carries out respectively filtering, discriminating and stored count.Last information process unit is according to the many groups measured value in several counting channels of the complementary matrix in spatial light modulator 2, hyperchannel counter 9 and the corresponding wavelength information of each passage, just can recover respectively the coloured image of visible ray and near infrared light, and can carry out component distributing analysis to interested wavelength, spectrogram that can also output wavelength-luminous power, and then obtain the characteristic peak of observed objects.

It should be noted that, in the time that utmost point low light level light intensity exceeds the investigative range of visible ray single-photon detector linear array 7 and near infrared light single-photon detector linear array 8, need on any light path position before single-photon detector linear array, suitable attenuator be set, for light decay being reduced to the investigative range of single-photon detector linear array, if utmost point low light level light intensity is in the investigative range of single-photon detector linear array, without attenuator is set again.

For feasibility and the practicality of checking native system algorithm, visible images and the infrared light image of in simulated experiment, choosing M17SWex nebula carry out R, G, B three primary colors decompose, carry out the light signal under the each wavelength channel of simulated spectra, be multiplied by respectively corresponding two-value random measurement matrix, the measurement Matrix Complementarity that wherein visible ray and infrared light image adopt, income value reloads additive white noise, count measurement value under the each wavelength channel that obtains simulating, in the situation that not knowing original image, hypothesis adopt method proposed by the invention to rebuild, recover the coloured image of visible ray and infrared light.In experiment, the resolution of all images is 300 × 600, adopt TV algorithm to carry out image reconstruction, obtain the result shown in Fig. 2, Fig. 3, wherein, Fig. 2 (a), Fig. 2 (b), Fig. 2 (c), Fig. 2 (d) are respectively R, G, B component and the colour originals of M17SWex nebula visible ray original image; Fig. 2 (e), Fig. 2 (f), Fig. 2 (g) are respectively the reconstruction figure of visible ray R, G, B component, and Fig. 2 (h) is the visible ray cromogram being finally inversed by according to this three-component; Fig. 3 (a), Fig. 3 (b), Fig. 3 (c), Fig. 3 (d) are respectively R, G, B component and the colour originals of M17SWex nebula infrared light original image; Fig. 3 (e), Fig. 3 (f), Fig. 3 (g) are respectively the reconstruction figure of infrared light R, G, B component, and Fig. 3 (h) is the infrared light cromogram being finally inversed by according to this three-component; From image quality, reconstruction coloured picture and former coloured picture difference are extremely low, thereby image quality is high, and application prospect is extensive.

It should be noted that, the former figure of M17SWex visible ray nebula is from Digitized Sky Survey (the Digitized Sky Survey of Britain's Schmidt telescope, be called for short DSS), the former figure of M17SWex infrared light nebula is from Spitzer space telescope, in infrared light image, we can clearly see that a dust cloud that seems imperial shape outwards flies out from once bright blast, but in visible images, this part dust cloud shrouds completely in shade, cannot differentiate.This imperial shape dust cloud that infrared light image presents is known as M17SWex, is probably the fixed star forming.These large-scale fixed stars have illuminated the nebula at the M17 of picture centre, also blown afloat one huge be positioned at M17 left hand edge form " foam " by luminous gas and dust.The gentle Galactic System through Sagittarius spiral arm just from right to left that is embodied in of the fixed star in this region, has caused " domino effect " of the milky way.The SWex dragon shape nebula of M17 is hidden in opaque dust cloud under visible ray.Be hidden in the region under " black clouds " and just can capture this with infrared ray, thereby disclose the situation of fixed star formation commitment.Infrared light is sightless in fact, is not have colouredly, and the color in figure identifies out, adopts equally three primary color theory, the light of red mark 24000nm, the light of green mark 8000nm, the light of blue mark 3600nm.If testing data is near infrared light, principle is the same.As can be seen from this embodiment, the present invention has very great help to disclosing astrophysics phenomenon, will in astronomical sight, ride on the crest of success.Embodiments of the invention not and be confined to this, all will not have good application at other as forward position scientific research fields such as light spectrum image-forming Measurement and analysis, biotic component detection, night vision observation, utmost point weak light detection, concentration detection, imaging of medical, military and national defense, monitoring, remotely sensed image, terrible imagings.

Finally it should be noted that, algorithm described in embodiment, through lot of experimental data checking, has been true and reliable, and collocation hardware just can be realized technical scheme of the present invention.Described all embodiment are only for to be further elaborated to the present invention, not absolute, can respective extension.Those of ordinary skill in the art should be appreciated that technical scheme of the present invention is modified, adds, deleted or is equal to replacement, and do not depart from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of claim scope of the present invention.

Claims (10)

1. the single photon spectrum of a complementary measurement counting imaging system, is characterized in that, described system comprises:

Optical imagery parts, spatial light modulator, first group of receipts optical alignment parts, second group of receipts optical alignment parts, first group of spectrum light splitting part, second group of spectrum light splitting part, visible ray single-photon detector linear array, near infrared light single-photon detector linear array, hyperchannel counter and information process unit; Wherein, the light that carries object under test information is imaged in described spatial light modulator by described optical imagery parts, and the emergent light after the random optical modulation of described spatial light modulator is divided into two arm direction transmission;

Receive optical alignment parts, first group of spectrum light splitting part and visible ray single-photon detector linear array and be arranged at the first arm direction, for surveying the photon of visible light, export some group pulse waveforms for described first group; Receive optical alignment parts, second group of spectrum light splitting part and near infrared light single-photon detector linear array and be arranged at the second arm direction, for surveying the photon of near infrared light spectrum, export some group pulse waveforms for described second group;

Described hyperchannel counter is arranged at via on the circuit after described visible ray single-photon detector linear array and near infrared light single-photon detector linear array, carries out filtering, discriminating and stored count for the pulse waveform under each passage;

Described information process unit is according to the corresponding wavelength information of each counting channel on the each count value on the complementary matrix in described spatial light modulator, hyperchannel counter and hyperchannel counter, use associated compressed sensing algorithm to reconstruct the coloured image of visible ray and near infrared light, and can be further analyzed the composition under specific wavelength.

2. the single photon spectrum of complementary measurement according to claim 1 counting imaging system, it is characterized in that, described visible ray single-photon detector linear array and near infrared light single-photon detector linear array are operated in by several corresponding different wave lengths the linear array that the single photon point probe under Geiger mode angular position digitizer forms, select respectively the single photon point detecting element of several corresponding different wave lengths for visible light wave range and near infrared light wave band, many group optical filters of can arranging in pairs or groups use;

Described visible ray single-photon detector linear array and near infrared light single-photon detector linear array position can phase double replacements;

The detection mode of described visible ray single-photon detector linear array and near infrared light single-photon detector linear array adopts fiber optic collimator or Free-space coupling;

Wherein, visible ray single-photon detector linear array and near infrared light single-photon detector linear array also can enable certain a line or a certain row obtain by avalanche photodiode arrays.

3. the single photon spectrum of complementary measurement according to claim 1 counting imaging system, it is characterized in that, when the every conversion of the stochastic matrix in spatial light modulator once, described hyperchannel counter records is once counted, the measured value in the compressed sensing algorithm that some batch total numerical value of its output adopt as information process unit.

4. the single photon spectrum of complementary measurement according to claim 1 counting imaging system, it is characterized in that, described first group of spectrum light splitting part and second group of spectrum light splitting part include: optical alignment part, light splitting part, measurement of angle part, photometric observation and measure portion;

Light absorbing part part and described optical alignment part are used in conjunction with, for the emergent light in a certain arm direction in spatial light modulator is all collected and preliminary collimation.

5. the single photon spectrum of complementary measurement according to claim 1 counting imaging system, it is characterized in that, in the time that utmost point low light level light intensity exceeds the investigative range of described visible ray single-photon detector linear array or near infrared light single-photon detector linear array, on any light path position before described visible ray single-photon detector linear array or near infrared light single-photon detector linear array, attenuator is set, for light decay being reduced to the investigative range of described single-photon detector linear array.

6. the single photon spectrum of a complementary measurement counting formation method, described method comprises:

Step 101) light that carries object under test information is imaged in spatial light modulator by optical imagery parts, and through the random optical modulation of described spatial light modulator, emergent light is divided into two arm directions with certain probability;

Step 102) respectively light beam is carried out light splitting and assembles and collect in two arm exit directions, be collected in visible ray single-photon detector linear array and near infrared light single-photon detector linear array, be recorded in hyperchannel counter with photon counting form;

Step 103) coloured image of realizing visible ray and near infrared light by compressed sensing algorithm rebuilds or correlation spectrum analysis, completes the single photon spectrum counting imaging of complementary measurement.

7. the single photon spectrum of complementary measurement according to claim 6 counting formation method, is characterized in that described step 103) further comprise:

Step 103-1), the step for compression sampling:

By spatial light modulator, the light being incident on it is carried out to random optical modulation, make two-way emergent light invest follow-up receipts optical alignment parts and single-photon detector linear array with certain probability respectively, the every conversion of random array in spatial light modulator once, measure simultaneously and record the corresponding total photon number of some specific wavelengths on two bands, be equivalent to the light intensity of some points on measure spectrum band, measure M time, M is much smaller than measured signal dimension N, to complete measured signal compression sampling in visible light wave range and near infrared light wave band;

Wherein, described spatial light modulator refers under the control of time dependent signal, the Real-Time Optical information processing device that can modulate photodistributed intensity on space, the light of the angle incident specifying by spatial light modulator can be reflexed to both direction with certain probability, what described random optical modulation referred to modulation matrix in spatial light modulator is very random, and this matrix adopts two-value to characterize the state of reflection direction;

Step 103-2), the step for sparse reconstruction:

Total photon number that visible ray single-photon detector linear array and near infrared light single-photon detector linear array are detected in twice transfer interval of spatial light modulator random array is as two groups of measured value y _iand y ' _i, corresponding stochastic matrix a respectively _ibenefit a ' with matrix _i, due to 0 and 1 two value representations for matrix element, the benefit of matrix is I-a, wherein I is unit matrix;

By a _iand a ' _ibe stretched to respectively a line, respectively as visible ray measure matrix A and near infrared light measure matrix A ' in i capable, measure M time, measure matrix all to amount to M capable for two, if visible ray single-photon detector linear array has P point, near infrared light single-photon detector linear array has Q point, the measured value of two arms has respectively P group and Q group, every group all amounts to M dimension, if two arms are respectively specifically chosen some groups of data, be more than or equal to three groups, utilize associated compressed sensing reconstruction algorithm to rebuild the coloured image of object of observation at visible light wave range and near infrared light wave band, wherein the coloured image of near infrared light wave band is marked, choose from big to small three kinds of infrared wavelengths by wavelength and be designated red green blue tricolor, get one group of data if appoint and just can carry out the constituent analysis of this wavelength.

8. the single photon spectrum of complementary measurement according to claim 6 counting formation method, it is characterized in that, the counting of each passage is added up, photon counting is converted into luminous power, in conjunction with the corresponding wavelength information of each passage, just can count the spectrogram of a wavelength-luminous power, and then obtain the characteristic peak of observed objects.

9. the single photon spectrum of complementary measurement according to claim 7 counting formation method, is characterized in that, the mathematical model of described compressed sensing is as follows:

Get a specific wavelength at visible-range and each of near infrared range, the signal under this two wavelength is stretched to row: sampling process can be regarded two groups of processes of measuring matrixes and two specific wavelength signals and doing inner product as, obtains respectively two groups of observation vector y _pi, y ' _qi, i=1,2 ..., M, p=1,2 ..., P, q=1,2 ..., Q, wherein system noise, if x be compressible or can sparse expression, wherein Ψ=[ψ ₁, ψ ₂..., ψ _n] be sparse transformation matrix, be associated degree of rarefication, so, the process of compression sampling can be described as following formula:

$[y_p,y_q^{\text{'}}]=[\mathrm{A\&Psi;}{\stackrel{\&OverBar;}{x}}_p+e,A^{\text{'}}\mathrm{\&Psi;}{\stackrel{\&OverBar;}{x}}_q^{\text{'}}+e^{\text{'}}]$

Wherein, A is set to two-value random measurement matrix;

Described sparse reconstruction is at known observation data y _p, y ' _qwith measurement matrix A, under the condition of A ', solve x _p, x ' _q, solving with the strategy of compressed sensing, specific algorithm is described as following formula:

$\underset{{\stackrel{\&OverBar;}{x}}_p}{\min }\frac{1}{2}{\left|\right|y_p-\mathrm{A\&Psi;}{\stackrel{\&OverBar;}{x}}_p\left|\right|}_2^2+\mathrm{\&tau;}{\left|\right|{\stackrel{\&OverBar;}{x}}_p\left|\right|}_1,\underset{{\stackrel{\&OverBar;}{x}}_p^{\text{'}}}{\min }\frac{1}{2}{\left|\right|y_q^{\text{'}}-A^{\text{'}}\mathrm{\&Psi;}{\stackrel{\&OverBar;}{x}}_q^{\text{'}}\left|\right|}_2^2+\mathrm{\&tau;}{\left|\right|{\stackrel{\&OverBar;}{x}}_q^{\text{'}}\left|\right|}_1$

Wherein, || ... || _lrepresent norm operator, only need the inferior measurement of M≤O (Klog (N/K)), just can be finally inversed by the light intensity space distribution information x of object under test under visible light wave range and two specific wavelengths of near infrared light wave band _pand x ' _qif, specifically choose separately 3 groups of measurement data at visible ray and near infrared light wave band, reuse above-mentioned model, just adopt three primary color theory can reconstruct the coloured image of visible light wave range and near infrared light wave band.

10. the single photon spectrum of complementary measurement according to claim 6 counting formation method, it is characterized in that, described compressed sensing algorithm is based on two measurement matrixes that meet complementary relationship, core algorithm strategy is regarded unit matrix to measure matrix participation computing as, described core algorithm is based on compressed sensing algorithm, after computing, add corrected parameter, the income value after correcting process again with x _pdo match operation, just obtain x ' _q, make visible light signal x _pwith near infrared light signal x ' _qcan obtain simultaneously.