CN108111759A - Towards the emulation design method of area array CCD opto-electronic conversion - Google Patents
Towards the emulation design method of area array CCD opto-electronic conversion Download PDFInfo
- Publication number
- CN108111759A CN108111759A CN201711412176.8A CN201711412176A CN108111759A CN 108111759 A CN108111759 A CN 108111759A CN 201711412176 A CN201711412176 A CN 201711412176A CN 108111759 A CN108111759 A CN 108111759A
- Authority
- CN
- China
- Prior art keywords
- ccd
- emulation
- opto
- output
- electronic conversion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/60—Noise processing, e.g. detecting, correcting, reducing or removing noise
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/60—Noise processing, e.g. detecting, correcting, reducing or removing noise
- H04N25/63—Noise processing, e.g. detecting, correcting, reducing or removing noise applied to dark current
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Transforming Light Signals Into Electric Signals (AREA)
Abstract
The present invention proposes a kind of emulation design method towards area array CCD opto-electronic conversion, including：According to the source spectral response curve of multispectral input picture, the CCD spectral responsivities of emulation wave band are obtained according to the emulation wavelength band simulation calculation of setting；CCD opto-electronic conversion models are established, and opto-electronic conversion, the voltage image file of output ccd detector response are completed by input of the wave band irradiation level field in incident detector face；Analog circuit model is established, carries out the analog circuit transmission process of pre-amplification, filtering and postposition amplification；Modulus quantitative model is established, and digital quantization processing is carried out to the voltage value data of output, is completed by the quantizing process of analog signal to digital signal and output digital image file.The present invention can better simulate the photoelectric conversion process of simulation different loads, and suitable for various full color imaging load, load design parameter reference can be provided for load design side.
Description
Technical field
The invention belongs to satellite remote sensing camera simulation technical fields, are related to a kind of emulation towards area array CCD opto-electronic conversion
Design method.
Background technology
Satellite imagery emulation is in design of satellites, predevelopment phase, for earth observation image quality under complicated image-forming condition
The important step of safeguards technique research.By links in analog satellite imaging process to the influence factor of image quality, emulation obtains
The analog image of satellite is taken, analysis of the user for satellite application ability can be supported, and then instruct the proposition of index demand；
It can aid in carrying out the analysis of satellite imagery quality.Before practical application, can in time it be diagnosed side by side by experimental simulation verification
Except hidden danger and failure, development quality can be effectively improved；In addition, studied by using emulation technology, it can also be significantly
Experiment number and quantity are reduced, so as to play the purpose for shortening the lead time, saving research fund, improving cost performance.For
The problem of new high-resolution, EO-1 hyperion, multiband earth observation satellite will be run into following imaging process, Yi Jiying
The factor of picture quality is rung, at present practical examples there is no to may be referred to.Therefore before the development of satellite remote sensing camera development work
Phase carries out the emulation of satellite imagery load opto-electronic conversion, helps intuitively, quantitatively to analyze the virtual condition in satellite transit
And picture quality.
The content of the invention
Problems to be solved by the invention are that being imaged the load design stage for remote sensing satellite can not accurate and visual acquisition phase
The problem of machine image quality.
Technological means for solving problem is to propose a kind of emulation design method towards area array CCD opto-electronic conversion,
Each effect stepwise factor in photoelectric conversion process is emulated, reference is provided for load design.
The present invention proposes a kind of emulation design method towards area array CCD opto-electronic conversion, includes the following steps：
Step 1, the source spectral response curve according to multispectral input picture are emulated according to the emulation wavelength band of setting
The CCD spectral responsivities of emulation wave band are calculated；
Step 2 is established CCD opto-electronic conversion models, and is completed by input of the wave band irradiation level field in incident detector face
Opto-electronic conversion, the voltage image file of output ccd detector response；
Step 3 establishes analog circuit model, and the voltage image file of CCD opto-electronic conversions model output is put in advance
Greatly, the analog circuit transmission process of filtering and postposition amplification, output voltage Value Data；
Step 4 establishes modulus quantitative model, and carries out digital quantization to the voltage value data of analog circuit model output
Processing, while considers the quantizing noise generated in quantizing process, complete by analog signal to digital signal quantizing process and
Output digital image file.
Further, as a preferred technical solution of the present invention, step 1 simulation calculation obtains emulation wave band
CCD spectral responsivities, including：
The emulation wavelength band of spectral transmittance and setting in the CCD spectral response curves of camera optical system,
Carry out the transmissivity of optical system extraction of emulation wave band；
According to the spectrum response of CCD and the emulation wavelength band set, complete to emulate the CCD responsivenesses of wave band
Extraction.
Further, as a preferred technical solution of the present invention, the step 2 is with the ripple in incident detector face
Opto-electronic conversion is completed in section irradiation level field for input, is specially：
Illumination image to be sampled is inputted, sampling obtains the illumination image that CCD is an actually-received；
According to the duty cycle of CCD components, the single illumination image visited member and be an actually-received is calculated；
According to the single responsiveness for visiting member and the time of integration, Integral Processing is carried out to single member of visiting；
According to the inconsistency response coefficient for visiting member, the single voltage for visiting member is calculated and exported；
According to CCD dynamic ranges to the single influence factor for visiting first output voltage, the first light of the single spies of CCD is preserved and exported
The transformed voltage image file of electricity.
Further, as a preferred technical solution of the present invention, the CCD dynamic ranges visit member output to single
The influence factor of voltage includes the noise of dark current, saturation output voltage and CCD.
Further, as a preferred technical solution of the present invention, the step 3 carries out voltage image file pre-
The analog circuit transmission process of amplification, filtering and postposition amplification, specially：
Impedance matching and amplification are carried out and according to the low of setting to the voltage image file of CCD opto-electronic conversions model output
Bandpass filter frequency exports analog voltage after carrying out low-pass filtering；
The gain amplification factor and biasing amplification quantity amplified according to the postposition of setting, the simulation exported to low-pass filtering
Voltage is adjusted and output voltage signal；
According to the circuit noise that camera noise emulation module generates, to the voltage signal superposed simulation of postposition amplification output
Circuit noise and output voltage Value Data.
Further, as a preferred technical solution of the present invention, low-pass filtering includes space filtering in the method
And frequency filtering.
Further, as a preferred technical solution of the present invention, the frequency filtering uses Butterworth low pass
Ripple device.
Further, as a preferred technical solution of the present invention, the step 4 exports analog circuit model
Voltage value data carries out digital quantization processing, is specially：
By the quantizing noise that camera noise emulation module generates be added to analog circuit model output voltage value data
On；
According to the quantitative model of foundation, the output voltage Value Data after the superposition is subjected to modulus quantization and output number
Word image file.
Invention effect is that emulation design method of the invention possesses following advantage：
(1), in the existing theoretical foundation of remote sensing camera photo-translating system, for the actual demand of project task, build
The remote sensing camera photo-translating system model of the detailed device parameters of the ccd detector that is based on, for CCD spectral response curves
Feature establishes spectral reflectivity model, and realizes software implementation, has expanded the function of photoelectronic imaging software, realizes multispectral figure
The emulation and processing of picture；Electrophotographic imaging process is refined, photoelectric conversion process is divided into four subsystems, is realized for each
The simulation study to Imaging of subsystem.
(2), compared, carried out for remote sensing camera light by voluntarily building experiment porch and emulating image in laboratory
The Validation of Simulation Models work of power conversion system, the experimental results showed that：Simulation software emulation degenerates the image with testing shooting
The similarity of the MTF rates of descent of image is 90%, it was demonstrated that simulation software model is correct and reliable.
Description of the drawings
Fig. 1 is the flow diagram of emulation design method of the present invention.
Fig. 2 is the spectral response curve figure of the present invention.
Fig. 3 is CCD opto-electronic conversions modeling and simulating process chart of the present invention.
Fig. 4 is the analog circuit modeling and simulating process chart of the present invention.
Fig. 5 quantifies modeling and simulating process chart for modulus of the present invention.
Fig. 6 is resampling process schematic of the present invention.
Fig. 7 divides schematic diagram for sampling area of the present invention.
Fig. 8 is the fundamental diagram of the low-pass filter the present invention is based on Fourier transform domain.
Specific embodiment
Hereinafter, it is described in detail based on attached drawing for the present invention.
As shown in Figure 1, the present invention proposes a kind of emulation design method towards area array CCD opto-electronic conversion, including as follows
Step：
Step 1, the source spectral response curve according to multispectral input picture are emulated according to the emulation wavelength band of setting
The CCD spectral responsivities of emulation wave band are calculated.
In the step 1, according to multispectral input picture source spectral response curve, as shown in Fig. 2, according to camera optics
The spectral response curve of the CCD of system, and according to emulation wavelength band, analyze the CCD spectral responsivities for emulating wave band, master
Want step as follows：
101) according to the spectral transmittance curve of camera optical system and emulation wavelength band, complete to emulate the light of wave band
System transmitance is extracted, and preserves into individual files.
102) according to the spectrum response of CCD and emulation wavelength band, the CCD responsivenesses for completing emulation wave band carry
It takes, preserves into individual files.
Step 2 is established CCD opto-electronic conversion models, and is completed by input of the wave band irradiation level field in incident detector face
Opto-electronic conversion, the voltage image file of output ccd detector response.
In the step 2, principle as shown in figure 3, CCD opto-electronic conversions according to time of integration of CCD, visit the ginsengs such as elemental size
Number, establishes CCD opto-electronic conversion models, using the wave band irradiation level field in incident detector face as input, completes opto-electronic conversion, defeated
Go out the voltage image file of ccd detector response, key step is as follows：
201) illumination image to be sampled is inputted, considers the influence of resolution ratio, by sampling process,
Obtain the illumination image that CCD is an actually-received；
202) according to the duty cycle of CCD components, the single illumination image visited member and be an actually-received is calculated；
203) according to the single responsiveness for visiting member and the time of integration, Integral Processing is carried out to single member of visiting；
204) according to the influence for the inconsistency response coefficient for visiting member, calculate and then export the single voltage Vi for visiting member.
205) finally according to CCD dynamic ranges to the single influence for visiting first output voltage, the influence of dynamic range is mainly wrapped
The noise of dark current, saturation output voltage and CCD is included, finally preserves the single voltage patterns for visiting first opto-electronic conversion output of output CCD
As file.
Step 3 establishes analog circuit model, and the voltage image file of CCD opto-electronic conversions model output is put in advance
Greatly, the analog circuit transmission process of filtering and postposition amplification, output voltage Value Data.
Its principle as shown in figure 4, the analog circuit modeling and simulating be used for simulate CCD output voltage image data into
The analog circuits transmission processes such as row pre-amplification, filtering and postposition amplification, in the links of artificial circuit, according to design electricity
The characterisitic parameter on road, response data of the emulation voltage data after these circuit devcies, key step are as follows：
301) according to the amplification factor on pre-arcing road, impedance matching is carried out to the analog voltage signal of ccd sensor output
And amplification.
302) according to the low-pass filter 3dB frequencies (Hz) of filter circuit, to the analog voltage signal of pre-arcing road output
After carrying out low-pass filtering, output output analog voltage.
303) according to the gain amplification factor of post amplifier and biasing amplification quantity, to filtering output analog voltage into
Row adjustment and output voltage signal.
304) circuit noise generated according to camera noise emulation module, to the voltage signal of postposition amplifying circuit output
Superposed simulation circuit noise and output voltage Value Data.
Preferably, low-pass filtering includes space filtering and frequency filtering in the method.The frequency filtering uses Bart
Butterworth low-pass filter.
Step 4 establishes modulus quantitative model, and carries out digital quantization to the voltage value data of analog circuit model output
Processing, while considers the quantizing noise generated in quantizing process, complete by analog signal to digital signal quantizing process and
Output digital image file.
In the step 4, principle is as shown in figure 5, modulus quantifies the voltage value number that modeling and simulating exports analog circuit
According to progress digital quantization processing, while consider the quantizing noise generated in quantizing process, complete by analog signal to digital signal
Quantizing process., key step is as follows：
401) by the quantizing noise that camera noise emulation module generates be added to analog circuit output voltage value data
On；
402) according to quantization maximum voltage, minimum voltage and the quantization digit of analog-digital converter, quantitative model is established；
403) according to quantitative model, the voltage value data result exported after superposition in step 401) is subjected to modulus quantization,
Complete quantizing process, output digital image file.
By more than satellite imagery load opto-electronic conversion simulation algorithm, the light of simulation different loads can better simulate
Electric transfer process suitable for various full color imaging load, can provide load design parameter reference for load design side.
2 to 8 pairs of specific embodiments of the invention are further described in detail below in conjunction with the accompanying drawings, wherein CCD light
Electricity conversion modeling and simulating process flow is as shown in Figure 3.Analog circuit modeling and simulating process flow is as shown in Figure 4.Modulus quantization is built
It is as shown in Figure 5 to imitate true process flow.
For spectral response simulation modeling：The responsiveness of CCD can use photoelectric current ILWith an area in t during integration
For AgThe signal charge number N that is accumulated of pixelsIt represents, form can be written as：
In formula, Φ is the radiancy of scenery.SIUnit represented with mA/W.
For CCD, frequently with FDA modes, therefore in practical application responsiveness can also be defined as pixel unit stream it is close
Spend output voltage V caused by σSSize, i.e.,：
Radiation energy current density is commonly used illumination (lx) in light measurement and is represented, using relational expression 1W/m2=20lx is changed
It calculates.
For multispectral input picture source, can its corresponding response be calculated according to its spectral band dividing condition subrane
Degree, it is assumed that spectral response curve is as shown in Figure 2.Data shown in figure and shown image curve all assume that value, have no actual object
Manage meaning.
For input spectrum, when input wavelength is λiWhen, corresponding spectral responsivity is Rλi, then for imaging
Wavelength band is in λi~λj, i<J, CCD spectral responsivity is：
It is mainly the optical signal being incident in photosensitive member to be converted to the process of electric signal that the opto-electronic conversion of CCD, which crosses j journeys,
Luminous flux on CCD photosurfaces is the incident total amount of (time for exposure) within a certain period of time；CCD is in imaging process simultaneously
With spatial sampling effect, main cause is as caused by the structure of CCD faces battle array, it is not continuous, but by small one by one
Face element composition, these small patches carry out spatial sampling to continuous incident ray, main in entire CCD photoelectric conversion processes
It will be there are three the stage.
First stage is that incident optical signal is converted to electric signal (electrons), and the electric field generated by electrode
It is collected；
Second stage is charge transfer process, and the electric signal that imaging area obtains is defeated according to certain timing control transfer
Go out；
Phase III is that signal reads amplification process, it is contemplated that the influence factor in photoelectric conversion process, photogenerated charge
(electrons) are converted to voltage signal, and amplify reading.
(1) CCD resamplings.Since CCD visits the impossible infinitesimal of elemental size, so can only be carried out successively to region scenery
Integration imaging, here it is spatial sampling processes.All it is discrete in view of the illumination image after illumination image to be sampled and sampling
Digital picture, sampling when will can regard the pixel of the illumination image of sampling as rectangular block rather than pixel.First calculate
Then coordinate position of each sampled points of CCD (taking pixel center) on Original Photo degree figure calculates the border of pixel after each sampling
Image pixel is repartitioned, as shown in Figure 6 in position.
Hereafter, the illumination being divided in same pixel is added up.It should be noted that due to CCD pixel size
It is not necessarily the integral multiple of original pixel, it is easy to the situation that original pixel is split by new pixel boundary occur.At this point, according to entrance
The area percentage of new pixel is calculated into the illumination size of sampled pixel.Here the new pixel after each sampling is divided into 9
Region, sampling area division schematic diagram are as shown in Figure 7.
The illumination image element that intermediate region is included is entire pixel, and illumination is directly included in the pixel after sampling.
And corner areas calculates illumination according to the original pixel area percentage enclosed into new pixel, and it is all fixed to calculate Central Plains picture element matrix length and width
Justice is 1, in this way, the coordinate of B1, B2, B3, B4 are integer.
Such as upper left corner area：
For another example upper middle region：
A1A2A3A4 encloses the total illumination of pixel after sampling：
E=EUpper left+EIn on+EUpper right+EIt is left+EIn+EIt is right+ELower-left+EIn under+EBottom right
(2) the actual illuminance that receives of member is visited to calculate.In CCD imaging processes, different devices has different duty cycles
(mark/space ratio, MSR), the then irradiation level that each CCD visits that member is an actually-received are：
Ereal=E × MSR
Note：ErealFor the illuminance actually received in flow chart, MSR is DutyRatio (the CCD duties in parameter list
Than).
(3) first time of integration is visited.The integration for visiting member is storage of the spatially discrete each photosensitive member to photogenerated charge
Journey, and stored in photosensitive member charge number depend on being incident on light intensity and the time of integration in the photosensitive member.Use mathematics
Expression formula is expressed as
Wherein, g (x, y) is the size of optical signal that CCD is received at pixel (x, y) position, and h (x ', y ', t) it represents
Arbitrary point on pixel (x, y) (x ', y ') optical signal is in the energy density of t moment, and T represents the time for exposure, and a, b then represent pixel
Length and width.
From the angle of the Realization of Simulation, for single band image, can be represented with following formula.
Wherein RλFor the response of corresponding spectral coverage, ErealλFor the irradiation level information that receives of actually practising physiognomy, τ is area array CCD
The time of integration.
(4) first heterogeneity is visited.In CCD manufacturing processes, the main reason for causing pixel Photo-Response Non-Uniformity, has：Device
Perimetric length can not in technique for the inconsistent and surface transparent electrode that is inconsistent, visiting first size of part substrate diffusion concentration
Accurate control etc..It shows as under uniform illumination, each member of visiting generates how much different charge is, ultimately results in each figure for visiting member
Image brightness is different.
The heterogeneity of response there is no unified definition, but a kind of more rigorous definition method is by the square of photoresponse
Root deviation and the mean ratio of response are used as the heterogeneity of CCD photoresponses.It has been generally acknowledged that photosensitive member is heterogeneous,
And CCD is approaches uniformity, also implies that transfer efficiency each time is the same.So have：
Here：
For average original response equivalent voltage；M is the total bit of line array CCD；VonFor n-th of photosensitive yuan of original response
Equivalent voltage.
Due to the presence of transfer indfficiency, the output signal V of CCDnPhotosensitive first original response V corresponding to itonNot phase
Deng.But according in front it is assumed that V can be calculated indirectlyonFor：
In formula, N is transfer number；ncpFor the number of phases of CCD.
This heterogeneity is generally characterized using linear function in the Realization of Simulation：
Y [i]=k [i] × x [i]+b [i]
Wherein, x [i] is i-th of voltage value for visiting member, and y [i] is the corresponding output for visiting member, and k [i] is i-th and visits member
Inconsistency response coefficient；B [i] is that i-th of inconsistency for visiting member biases coefficient.It can generally be given in the device handbook of CCD
Light extraction response nonuniformity (PRNU) and fixed pattern noise (FPN), wherein b [i] are the inconsistency of spectral response, k
[i] is PRNU-FPN i.e. distribution intensity, the distribution Normal Distribution of the two parameters, standard deviation mainly by FPN and
PRNU is determined.
(5) device noise.Photon noise, shot noise are primarily present in CCD photoelectric conversion processes, fertile zero noise, is turned
Move noise, dark current noise and output noise；Analogue noise is primarily present in analog circuit simulation process；Quantify in modulus
Quantizing noise is primarily present in the process.Several main noises present in CCD opto-electronic conversions：
1) photon noise：The emission process of photon is random, and therefore, CCD is equally also considered as in collection optical signal charge
As soon as being a random process, such random process constitutes a kind of noise source, this noise source is determined by the property of photon
's.This noise can be more serious when low-light (level) images.
2) shot noise：The presence of shot noise is due to discrete and quantization the ripple at light or electric current during exercise
Bag is formed.After the photogenerated charge number that the unit interval generates has slight difference compared to average value, that is, result in shot noise
It generates.According to the characteristic of white noise, shot noise is not related with frequency, it in all frequency ranges power distribution all
It is uniform.Illumination is relatively low, in the environment of contrast unobvious, in other noise informations after processing of circuit is suppressed, dissipates
Grain noise becomes the main noise information in CCD imaging processes, it determines the limit noise level of CCD device.
3) fertile zero noise：Supplement potential well position using fertile zero charge, enable signal charge by random region into
Row transfer, can drop charge and be divided into electronics fertilizer zero-sum optics fertilizer zero.Therefore the noise generated is also classified into fertile zero noise of electronics and light
Fertile zero noise is learned, optics zero noise of fertilizer is mainly determined that electronics zero noise of fertilizer is then by electron injection by the size of the bias lighting of CCD
Fertile zero mechanism determines.
4) noise is shifted：In CCD transfer processes, previous charge packet not yet shifts completely, and a part of charge remains in gesture
In trap, so as to form noise jamming to latter charge packet.Cause transfer noise basic reason it is main there are three：Interfacial state is captureed
It obtains, transfer indfficiency and figure capture.
5) dark current noise：The main reason for dark current generates is the current-carrying generated inside semiconductor due to warm-up movement
Son is added in potential well, is then transferred under the action of driving pulse, and output current.Even if dark current is in no light
In the case of there is also.Dark current is divided into diffusion dark current and surface dark current etc..
Diffusion dark current is that the free space under conducting channel and potential well by CCD generates, and its diffusion length is got over
Short, potential well number is more, and dark current is bigger.
Surface dark current is to be jumped to jump to conduction band from interfacial state under thermal excitation by electronics, again by gesture after generation free electron
The electric current that trap is eventually formed as dark charge reception.
All ccd detectors can all be influenced be subject to dark current, it can determine sensitivity and the dynamic range of device.
Since the size of temperature determines the size of the dark current noise generated during warm-up movement, and temperature often increases by 5 DEG C~6
DEG C, dark current will be increased to original twice.
Dark current is also related with the time that charge packet stores in potential well, and storage time is longer, the dark current noise of generation
Also it is bigger.In weak signal conditions, when CCD by the way of long-time integration when being imaged, dark current noise will be main
Influence the principal element of image quality.
In addition, in CCD, local lattice is likely to cause dark current peak there are magazine or existing defects.With mixing
Miscellaneous concentration increases, and nearer from surface distance, electric field strength is accordingly bigger.From surface most nearby, electric field strength reaches maximum,
Dark current peak value is more susceptible to.Dark current peak value can cause image background larger fluctuation occur.
6) output noise：Ccd signal is that the signal charge of CCD is converted into corresponding voltage by floating capacity is right
It exports, and is mostly exported by the way of floating diffusion capacitance afterwards.
In above-mentioned noise, since dark current noise, transfer noise have contemplated that in the section of opto-electronic conversion one, so connecing
In the modeling scheme to get off, dark current is not just considered further that and shifts the influence of noise, mainly considers shot noise, output is made an uproar
The influence of sound.
For more than noise, to classify in simulation process to different noises, be built according to different noise types
Vertical different noise model.CCD is sampled to be inevitable with the noise introduced in photoelectric conversion process, the noise that CCD is introduced
It is the main source of system noise, even more influences the principal element of system imaging quality.In the Realization of Simulation, due to making an uproar for CCD
Sound is difficult to be quantified on a microscopic level, these noises is analyzed one by one without suitable noise model, so simulation process
It is middle to need to replace the noise in microcosmic point using quantitative model, they are equivalent to a white noise for being superimposed upon circuit end
Sound.
Since the noise majority generated in CCD photoelectric conversion processes belongs to white Gaussian noise, the research of this paper and after
It it will be assumed that all noises obey the Gaussian Profile that average is μ, variance is σ in photoelectric conversion process in continuous emulation：
According to Gaussian noise model, in the actual signal of the Gaussian noise superposition that model is generated to the end：
Vreal=V+Vnoise
The main noise generated in circuit has dark current noise, output noise and reset noise etc..Also due to do not have
There is suitable noise model to analyze these noises one by one, it is therefore assumed that noise all Normal Distributions that circuit module generates
Model, probability density function are：
Wherein μ and σ is the statistical value of noise, and μ is the average value of equivalent output noise voltage, is
σ is then the root-mean-square value of noise voltage fluctuation at any time, is
Generally, being imaged the artificial circuit part of load opto-electronic conversion mainly includes four modules：Pre-amplifier module,
Low pass filter blocks, operational amplifier module and correlated double sampling circuit module.Wherein correlated double sampling circuit is main
Effect is for eliminating reset noise, extracting useful signal, and is transferred to modulus sample circuit.In emulation module, by
It is excessive in the factor for influencing noise, noise is reduced to normal distribution model, there is no addition reset noise information, so
It can be without considering correlated double sampling circuit during modeling.Therefore in the Realization of Simulation, analog circuit can be divided into three parts
As pre-amplification circuit, low-pass filter circuit and postposition amplifying circuit.Analog circuit emulation is exactly to emulate above-mentioned each ring
Save the response to signal.
According to the characterisitic parameter of circuit, response data of the emulation signal after these circuit modules.
(1) pre-amplification circuit and postposition amplifying circuit.The main function of pre-amplifier module is to receive optical signal simultaneously
Optical signal is converted into electric signal, then electric signal is amplified, then by amplified electric signal transmission to signal below
In processing system.Pre-amplification circuit is very important in entire photoelectric conversion process, is primarily due to preposition amplification electricity
For the optical signal that road receives after amplifying circuit amplifies, signal is all very faint, it is easy to is submerged in noise, therefore
Its performance quality directly determines the quality of the performance of entire circuit system.In the Realization of Simulation, will can entirely it amplify
Circuit is expressed as shown in following formula.
V1=V0×B1
V in formula0For initial input voltage image, V1For the voltage after preventing big circuit, B in above formula1It represents pre-
Amplification factor.
Postposition amplifying circuit can be expressed as following formula in the Realization of Simulation.
V3=V2×B2+A
V2For the voltage after low-pass filtering, V2With gain amplification factor B2It is multiplied along with biasing amplification quantity A is
Last analog circuit simulation data result V3。
(2) low-pass filter, general relatively common filtering method have space filtering and frequency filtering two ways.It is empty
Between filtering refer to then directly making spatial variations to each pixel of image in domain is handled, frequency filtering then refers to image
Information goes in frequency domain after Fourier changes and spectrum component is handled.For two kinds of processing modes, space
Domain filtering is a kind of neighborhood operation, and frequency filtering is then more directly perceived.The key issue of frequency filtering is to select filtering
Device transmission function, founding mathematical models pass through inverse Fourier transform finally by by the image information that completion is handled in frequency domain,
The output image that finally obtains that treated.
The theoretical foundation of filter in spatial domain and frequency filtering is all convolution theorem.
The expression formula at formula both ends is one group of Fourier transform pairs, also this means that the convolution of two spaces function can be with
The inverse transformation of the product changed by Fourier obtains.That is, in the spatial domain, image f (x, y) to be filtered and filtering
The convolution algorithm of mask h (x, y) can be multiplied by H (u, v) by F (u, v) in frequency domain and be calculated.
Being filtered in frequency domain can regard as using a filter transfer function to change F (u, v), then
Calculate the inverse Fourier transform of H (u, v) F (u, v), it is possible to accordingly obtain the image of filter in spatial domain.Its principle such as Fig. 8 institutes
Show.
The flat amplitude response of Butterworth (Butterworth) wave filter is maximum, is widely used in the communications field.Phase
Than in Bezier (bessl), Chebyshev (chebyshev) wave filter, Butterworth filtering has in linear phase, attenuation
Balanced advantage in terms of slope and loading characteristic so that Butterworth LPF is more suitable for image procossing.
For n rank Butterworth LPFs, this wave filter is non-sharp by transmission function is：
In formula, D0Be by frequency, whereinAs D (u, v)=D0When,N is
Filter order is positive integer, is mainly used to control the rate of decay of frequency.
In general, generally being selected by frequency for low-pass filter makes H (u, v) drop to itThe frequency values at place, this
When transmission function can be written as：
As n=1, Butterworth LPF can improve the contrast of image without " ring " phenomenon；Work as n
When=2, it may appear that faint " ring " phenomenon；And when n is bigger, " ring " phenomenon is then more apparent, at this time also closer to preferable
Low-pass filter.
Finally, imaging load modulus quantifies modulus and quantifies voltage value data progress number of the emulation to analog circuit output
Quantification treatment, while consider the quantizing noise generated in quantizing process, it completes by the quantization of analog signal to digital signal
Journey.
Modulus quantifies emulation and carries out digital quantization processing to the voltage value data of analog circuit output, according to analog-to-digital conversion
Quantization maximum voltage, minimum voltage and the quantization digit of device, establish quantitative model；
Wherein, N is quantization digit；VmaxAnd VminThe minimum and maximum signal voltage respectively exported.
To sum up, the present invention is compared by voluntarily building experiment porch in laboratory with emulating image, has been carried out for remote sensing
The Validation of Simulation Models work of camera photo-translating system, the experimental results showed that：Simulation software emulates the degenerate image and reality
The similarity for testing the MTF rates of descent of captured image is 90%, it was demonstrated that simulation software model is correct and reliable.
It should be noted that described above is only the preferred embodiment of the present invention, it should be understood that for this field skill
For art personnel, several changes and improvements can also be made on the premise of the technology of the present invention design is not departed from, these all include
Within the scope of the present invention.
Claims (8)
1. a kind of emulation design method towards area array CCD opto-electronic conversion, which is characterized in that include the following steps：
Step 1, the source spectral response curve according to multispectral input picture are obtained according to the emulation wavelength band simulation calculation of setting
To the CCD spectral responsivities of emulation wave band；
Step 2 establishes CCD opto-electronic conversion models, and completes photoelectricity by input of the wave band irradiation level field in incident detector face
Conversion, the voltage image file of output ccd detector response；
Step 3 establishes analog circuit model, and carries out pre-amplification, filter to the voltage image file of CCD opto-electronic conversions model output
Ripple and the analog circuit transmission process of postposition amplification, output voltage Value Data；
Step 4 establishes modulus quantitative model, and carries out digital quantization processing to the voltage value data of analog circuit model output,
Consider the quantizing noise generated in quantizing process simultaneously, complete by the quantizing process of analog signal to digital signal and output number
Word image file.
2. the emulation design method according to claim 1 towards area array CCD opto-electronic conversion, which is characterized in that the step
Rapid 1 simulation calculation obtains the CCD spectral responsivities of emulation wave band, including：
The emulation wavelength band of spectral transmittance and setting in the CCD spectral response curves of camera optical system, carries out
Emulate the transmissivity of optical system extraction of wave band；
According to the spectrum response of CCD and the emulation wavelength band set, the CCD responsivenesses extraction of emulation wave band is completed.
3. the emulation design method according to claim 1 towards area array CCD opto-electronic conversion, which is characterized in that the step
Rapid 2 complete opto-electronic conversion by input of the wave band irradiation level field in incident detector face, are specially：
Illumination image to be sampled is inputted, sampling obtains the illumination image that CCD is an actually-received；
According to the duty cycle of CCD components, the single illumination image visited member and be an actually-received is calculated；
According to the single responsiveness for visiting member and the time of integration, Integral Processing is carried out to single member of visiting；
According to the inconsistency response coefficient for visiting member, the single voltage for visiting member is calculated and exported；
According to CCD dynamic ranges to the single influence factor for visiting first output voltage, the first opto-electronic conversion of the single spies of CCD is preserved and exported
Voltage image file afterwards.
4. the emulation design method according to claim 3 towards area array CCD opto-electronic conversion, which is characterized in that the CCD
Dynamic range includes the single influence factor for visiting first output voltage the noise of dark current, saturation output voltage and CCD.
5. the emulation design method according to claim 1 towards area array CCD opto-electronic conversion, which is characterized in that the step
Rapid 3 pairs of voltage image files carry out the analog circuit transmission process of pre-amplification, filtering and postposition amplification, are specially：
Impedance matching and amplification and the low pass filtered according to setting are carried out to the voltage image file of CCD opto-electronic conversions model output
Wave frequency rate exports analog voltage after carrying out low-pass filtering；
According to the postposition of setting amplify gain amplification factor and biasing amplification quantity, to the analog voltage that low-pass filtering is exported into
Row adjustment and output voltage signal；
According to the circuit noise that camera noise emulation module generates, make an uproar to the voltage signal superposed simulation circuit of postposition amplification output
Sound and output voltage Value Data.
6. the emulation design method according to claim 5 towards area array CCD opto-electronic conversion, which is characterized in that the side
Low-pass filtering includes space filtering and frequency filtering in method.
7. the emulation design method according to claim 6 towards area array CCD opto-electronic conversion, which is characterized in that the frequency
Rate filtering is using Butterworth LPF.
8. the emulation design method according to claim 1 towards area array CCD opto-electronic conversion, which is characterized in that the step
The voltage value data of rapid 4 pairs of analog circuit models output carries out digital quantization processing, is specially：
The quantizing noise that camera noise emulation module generates is added in the voltage value data of analog circuit model output；
According to the quantitative model of foundation, the output voltage Value Data after the superposition is subjected to modulus quantization and output digitized map
As file.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711412176.8A CN108111759A (en) | 2017-12-23 | 2017-12-23 | Towards the emulation design method of area array CCD opto-electronic conversion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711412176.8A CN108111759A (en) | 2017-12-23 | 2017-12-23 | Towards the emulation design method of area array CCD opto-electronic conversion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108111759A true CN108111759A (en) | 2018-06-01 |
Family
ID=62212582
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711412176.8A Pending CN108111759A (en) | 2017-12-23 | 2017-12-23 | Towards the emulation design method of area array CCD opto-electronic conversion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108111759A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114609073A (en) * | 2022-05-10 | 2022-06-10 | 安徽中科谱康科技有限公司 | High-intensity spectrum measuring method and system and spectrum measuring device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1134898A2 (en) * | 2000-03-14 | 2001-09-19 | Lucent Technologies Inc. | Carrier-dependent dithering for analog-to-digital conversion |
| JP2007272666A (en) * | 2006-03-31 | 2007-10-18 | Matsushita Electric Ind Co Ltd | Image evaluation method and image evaluation device |
| CN101120507A (en) * | 2005-02-15 | 2008-02-06 | 皇家飞利浦电子股份有限公司 | AD converter arrangement |
| CN203849164U (en) * | 2013-06-14 | 2014-09-24 | 苏州耀宏光电科技有限公司 | Light path structure of spectrum transmittance measuring system |
| CN105049841A (en) * | 2015-07-24 | 2015-11-11 | 中国科学院合肥物质科学研究院 | Method for enhancing color displaying capability of color camera through single-channel pre-optical filter |
| CN106650049A (en) * | 2016-12-05 | 2017-05-10 | 北京空间机电研究所 | Time-division dynamic imaging simulation method for static rail area array remote sensing camera |
| CN106841086A (en) * | 2016-11-22 | 2017-06-13 | 北京空间机电研究所 | A kind of method for improving Atmospheric Survey Fourier spectrometer signal to noise ratio |
| CN107092752A (en) * | 2017-04-24 | 2017-08-25 | 北京理工大学 | A kind of optical camera simulation imaging method and system based on ray tracing |
-
2017
- 2017-12-23 CN CN201711412176.8A patent/CN108111759A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1134898A2 (en) * | 2000-03-14 | 2001-09-19 | Lucent Technologies Inc. | Carrier-dependent dithering for analog-to-digital conversion |
| CN101120507A (en) * | 2005-02-15 | 2008-02-06 | 皇家飞利浦电子股份有限公司 | AD converter arrangement |
| JP2007272666A (en) * | 2006-03-31 | 2007-10-18 | Matsushita Electric Ind Co Ltd | Image evaluation method and image evaluation device |
| CN203849164U (en) * | 2013-06-14 | 2014-09-24 | 苏州耀宏光电科技有限公司 | Light path structure of spectrum transmittance measuring system |
| CN105049841A (en) * | 2015-07-24 | 2015-11-11 | 中国科学院合肥物质科学研究院 | Method for enhancing color displaying capability of color camera through single-channel pre-optical filter |
| CN106841086A (en) * | 2016-11-22 | 2017-06-13 | 北京空间机电研究所 | A kind of method for improving Atmospheric Survey Fourier spectrometer signal to noise ratio |
| CN106650049A (en) * | 2016-12-05 | 2017-05-10 | 北京空间机电研究所 | Time-division dynamic imaging simulation method for static rail area array remote sensing camera |
| CN107092752A (en) * | 2017-04-24 | 2017-08-25 | 北京理工大学 | A kind of optical camera simulation imaging method and system based on ray tracing |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114609073A (en) * | 2022-05-10 | 2022-06-10 | 安徽中科谱康科技有限公司 | High-intensity spectrum measuring method and system and spectrum measuring device |
| CN114609073B (en) * | 2022-05-10 | 2022-07-29 | 安徽中科谱康科技有限公司 | High-intensity spectrum measuring method and system and spectrum measuring device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107092752B (en) | A kind of optical camera simulation imaging method and system based on ray tracing | |
| KR100850729B1 (en) | Method and apparatus for enhancing data resolution | |
| CN106871925B (en) | A kind of remote sensing satellite relative radiometric calibration processing method of in-orbit comprehensive dynamic adjustment | |
| Brusa et al. | Increasing the precision of canopy closure estimates from hemispherical photography: Blue channel analysis and under-exposure | |
| EP1667435A2 (en) | Sensing apparatus, method for sensing energy field and camera | |
| CN109974854B (en) | Radiation correction method for frame-type FPI (field programmable Gate array) hyperspectral image | |
| CN102625059B (en) | Dynamic range extension for CMOS image sensors for mobile applications | |
| CN101635782B (en) | Method and device for obtaining image based on dynamic time delay integral | |
| Chen et al. | Digital camera imaging system simulation | |
| WO2007069736A1 (en) | Method of evaluating tree body production power, imaging device for evaluating tree body production power and program for evaluating tree body production power | |
| CN111157225A (en) | Labview-based EMCCD chip full-performance parameter testing method | |
| CN113447118B (en) | Multispectral imaging chip capable of realizing color imaging and color imaging method | |
| Lang et al. | Digital photography for tracking the phenology of an evergreen conifer stand | |
| CN112461360B (en) | High-resolution single photon imaging method and system combined with physical noise model | |
| CN115641270A (en) | High dynamic range expansion method suitable for SCMOS (Single chip multiple Access) dual-gain imaging system | |
| Hanji et al. | Noise-aware merging of high dynamic range image stacks without camera calibration | |
| CN108111759A (en) | Towards the emulation design method of area array CCD opto-electronic conversion | |
| CN113256733A (en) | Camera spectral sensitivity reconstruction method based on confidence voting convolutional neural network | |
| CN106550199B (en) | Multilayer high dynamic range sensor | |
| CN104660929B (en) | A kind of voltage integrating meter type cmos image sensor exported using voltage integrating meter | |
| CN106550197A (en) | The incident light on an imaging sensor of modulation | |
| Yang et al. | On pixel detection threshold in the gigavision camera | |
| CN108154478B (en) | A kind of remote sensing image processing method | |
| Zhang et al. | Signal-to-noise analysis of point target detection using image pixel binning for space-based infrared electro-optical systems | |
| Schöberl et al. | Building a high dynamic range video sensor with spatially nonregular optical filtering |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180601 |