CN103884427A - Hyperspectral and high-spatial-resolution image obtaining method and device - Google Patents

Abstract

The invention relates to the technical field of image and video processing technologies and discloses a hyperspectral and high-spatial-resolution image obtaining method and device. The method comprises the steps of obtaining an RGB image with high spatial resolution and a hyperspectral grey image, wherein the RGB image is a whole area sampling image, the grey image is a partial sampling image passing through an evenly-distributed rectangular hole mask; enabling the RBG image to be aligned at and stacked with the grey image so as to obtain a comprehensive image; for pixels with sampling information in the grey image on the combined image, directly obtaining spectral information of the pixels from the grey image; for pixels with sampling information shortage in the grey image on the combined image, calculating the spectral information of the pixels through the spectrum spread algorithm based on double-edge filter. The hyperspectral and high-spatial-resolution image obtaining method and device can be used for obtaining hyperspectral and high-spatial-resolution images, and have the advantages of being simple and convenient in algorithm, real-time and quick.

Description

The image acquiring method of high spectrum high spatial resolution and device

Technical field

The present invention relates to image/video processing technology field, be specifically related to a kind of image acquiring method and device of high spectrum high spatial resolution.

Background technology

Most imaging camera work all records for scene image based on red, green, blue three look information, although three look sensing imaging techniques meet human visual system's imaging demand, but from the angle of physical principle, reality scene not only has three look information so simple.Actual light source is sent or is had abundant wavelength through the scene light of object reflection, wherein visible ray covers from 390nm until the extensive region of 780nm, in this section of wavelength coverage, the distribution of light light intensity has comprised a large amount of information, can effectively reflect the natural quality of light source, object and scene.

The triple channel collection of tradition color camera can be regarded as light spectrum carries out the result of integration amount on the integrated curve of the sensor response of corresponding three kinds of colors, but the process of this integration has been brought a large amount of loss of spectral information just.So correspondingly, spectral measurement is conceived to obtain the spectral information of losing in this process.

Early stage spectrometer, as the Image-forming instrument based on spacescan or spectral scan, focuses on the resolution that improves spectral information, but is limited to the seizure to single beam line.In recent years, the scientific research of spectra collection has just further realized the breakthrough in temporal resolution, emerge as based on tomoscan, based on coding aperture and based on multiple hyper-spectral data gathering technology such as mask prismatic decompositions, can complete the collection for high-resolution spectrum video, establish know-why basis for mating industrial or agricultural practical application and scientific research demand.

By optical imagery element, spatial sampling device (as mask, slit etc., the present invention is introduced as an example of mask example), the front-end system of the formation such as light-splitting device (prism, grating etc., the present invention is introduced for example with prism), sensor (the present invention is introduced for example with gray scale camera) can by the shooting to object obtain the object of being clapped reduction spatial resolution, comprise the video that enriches spectral information (high spectral resolution).Use light-dividing device to make RGB camera capture identical scene with gray scale camera, can obtain the video of high spatial resolution, low spectral resolution.Fusion is aimed at and calculated to two-path video in real time, just can obtain the video of high spatial resolution, high spectral resolution.How to calculate quickly and accurately fusion, remain a difficult problem.Therefore, the research of this respect has important and using value widely.

Summary of the invention

The present invention is intended at least solve the difficult problem that the high spectral resolution that exists in prior art and high spatial resolution are difficult to obtain simultaneously.

For this reason, one object of the present invention is to propose a kind of image acquiring method of high spectrum high spatial resolution.

Another object of the present invention is to propose a kind of image acquiring device of high spectrum high spatial resolution.

To achieve these goals, according to the image acquiring method of the high spectrum high spatial resolution of the embodiment of one aspect of the invention, comprise the following steps: the RGB figure of high spatial resolution and the gray-scale map of high spectral resolution that obtain Same Scene, wherein said RGB figure is region-wide sampled images, and described gray-scale map is the part sampled images through being uniformly distributed rectangle hole mask; Described RGB figure is alignd and superposeed with described gray-scale map, obtain synthetic image; For pixel on described synthetic image, have sample information in described gray-scale map, directly from described gray-scale map, obtain the spectral information of described pixel; For the pixel of sample information disappearance on described synthetic image, in described gray-scale map, the spectrum propagation algorithm of application based on bilateral filtering calculates the spectral information of described pixel.

According to the image acquiring method of the high spectrum high spatial resolution of the embodiment of the present invention, after obtaining the synthetic image after RGB figure and gray-scale map alignment, synthetic image is applied to the spectrum propagation algorithm based on bilateral filtering, obtain the spectral information of each pixel.The method has overcome light spectrum image-forming field and has been difficult to obtain high spatial resolution and a high spectral resolution difficult problem simultaneously, and the development of light spectrum image-forming is had to great importance; Needed equipment is comparatively simple, and cost is little, after integrated, is convenient to carry out commercial production.

In addition, the image acquiring method of the high spectrum high spatial resolution of the embodiment of the present invention also has following additional technical feature:

In one embodiment of the invention, described for pixel on described synthetic image, have sample information in described gray-scale map, the spectral information that directly obtains described pixel from described gray-scale map specifically comprises: the described pixel on described gray-scale map is carried out to spectral evolution, obtain the brightness value corresponding to light of different wave length.

In one embodiment of the invention, the described pixel for sample information disappearance on described synthetic image, in described gray-scale map, the spectral information that the spectrum propagation algorithm of application based on bilateral filtering calculates described pixel specifically comprises: remember that the object pixel point coordinate that need to ask for spectral information on described synthetic image is (i, j), with (i, j) for the center of circle, paint circle take preset reference distance R as radius, note border circular areas is Ψ, detects and in Ψ, has the reference image of at least one known spectra information vegetarian refreshments k; Calculate the spectral information of described target pixel points according to following formula,

${\mathrm{ms}}_{\mathrm{ij}}\left(\mathrm{\λ}\right)=\underset{c\∈R,G,B}{\mathrm{\Σ}}\frac{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})\·{\mathrm{\ρ}}_k^c\·{\mathrm{ms}}_k^c\left(\mathrm{\λ}\right)\·r_{\mathrm{\λ}}^c}{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})}$

Wherein, ms _ij(λ) brightness value of the light that the wavelength at expression target pixel points place is λ,

be illustrated in poor at the pixel value of c passage of target pixel points described in RGB figure and reference image vegetarian refreshments k, d _krepresent the theorem in Euclid space distance of described target pixel points and reference image vegetarian refreshments k, σ _rrepresent the first default Gauss operator standard deviation, σ _srepresent the second default Gauss operator standard deviation, represent the brightness ratio of described target pixel points and reference image vegetarian refreshments k c passage in RGB figure

the brightness value of the light that the wavelength that represents reference image vegetarian refreshments k place is λ,

represent that the light that wavelength is λ accounts for its ratio in R, G, the three-channel response ratio of B sum in the response ratio of c passage.

In one embodiment of the invention, calculate the spectral information of target pixel points take GPU as hardware.

The image acquiring device of the high spectrum high spatial resolution of embodiment according to a further aspect of the invention, comprising: RGB figure acquisition module, and for obtaining the RGB figure of high spatial resolution, wherein said RGB figure is region-wide sampled images; Gray-scale map acquisition module, for obtaining the gray-scale map of high spectral resolution, wherein said gray-scale map is the part sampled images through being uniformly distributed rectangle hole mask; Alignment laminating module, described alignment laminating module is connected with gray-scale map acquisition module with described RGB figure acquisition module respectively, for described RGB figure is alignd and superposeed with described gray-scale map, obtains synthetic image; Spectral information computing module, described spectral information computing module is connected with the described laminating module that aligns, for the pixel for sample information disappearance on described synthetic image, described gray-scale map, the spectrum propagation algorithm of application based on bilateral filtering calculates the spectral information of described pixel; Output module, described output module is connected with spectral information computing module with the described laminating module that aligns respectively, for pixel on described synthetic image, have sample information in described gray-scale map, described load module reads spectral information the output of described pixel from described alignment laminating module, for the pixel of sample information disappearance on described synthetic image, in described gray-scale map, described load module reads spectral information result of calculation the output of described pixel from described spectral information computing module.

According to the image acquiring device of the high spectrum high spatial resolution of the embodiment of the present invention, after the synthetic image obtaining after RGB figure and gray-scale map alignment, synthetic image is applied to the spectrum propagation algorithm based on bilateral filtering, obtain the spectral information of each pixel.This device has overcome light spectrum image-forming field and has been difficult to obtain high spatial resolution and a high spectral resolution difficult problem simultaneously, and the development of light spectrum image-forming is had to great importance; Needed equipment is comparatively simple, and cost is little, after integrated, is convenient to carry out commercial production.

In addition, also there is following additional technical feature according to the image acquiring device of the high spectrum high spatial resolution of the embodiment of the present invention:

In one embodiment of the invention, described for pixel on described synthetic image, have sample information in described gray-scale map, the spectral information that directly reads described pixel from described alignment laminating module specifically comprises: the described pixel described gray-scale map is carried out to spectral evolution, read the brightness value corresponding to light of different wave length.

In one embodiment of the invention, described spectral information computing module specifically comprises: reference zone defines submodule, be (i for the object pixel point coordinate that need to ask for spectral information defining on described synthetic image, j) reference zone, with (i, j) for the center of circle, paint circle take preset reference distance R as radius, note border circular areas is Ψ, detecting has the reference image of at least one known spectra information vegetarian refreshments k in Ψ; Calculating sub module, described calculating sub module is used for calculating according to following formula the spectral information of described target pixel points, ${\mathrm{ms}}_{\mathrm{ij}}\left(\mathrm{\λ}\right)=\underset{c\∈R,G,B}{\mathrm{\Σ}}\frac{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})\·{\mathrm{\ρ}}_k^c\·{\mathrm{ms}}_k^c\left(\mathrm{\λ}\right)\·r_{\mathrm{\λ}}^c}{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})},$ Wherein, ms _ij(λ) brightness value of the light that the wavelength at expression target pixel points place is λ,

be illustrated in poor at the pixel value of c passage of target pixel points described in RGB figure and reference image vegetarian refreshments k, d _krepresent the theorem in Euclid space distance of described target pixel points and reference image vegetarian refreshments k, σ _rrepresent the first default Gauss operator standard deviation, σ _srepresent the second default Gauss operator standard deviation,

represent the brightness ratio of described target pixel points and reference image vegetarian refreshments k c passage in RGB figure

the brightness value of the light that the wavelength that represents reference image vegetarian refreshments k place is λ,

represent that the light that wavelength is λ accounts for its ratio in R, G, the three-channel response ratio of B sum in the response ratio of c passage.

In one embodiment of the invention, in described calculating sub module, calculate the spectral information of target pixel points take GPU as hardware.

Additional aspect of the present invention and advantage in the following description part provide, and part will become obviously from the following description, or recognize by practice of the present invention.

Accompanying drawing explanation

Above-mentioned and/or additional aspect of the present invention and advantage accompanying drawing below combination is understood becoming the description of embodiment obviously and easily, wherein:

Fig. 1 is the process flow diagram of the image acquiring method of the high spectrum high spatial resolution of the embodiment of the present invention.

Fig. 2 a and Fig. 2 b are respectively middle RGB figure and the gray-scale maps of embodiment of the present invention Same Scene.

Fig. 3 is the spectral information schematic diagram that the single sample point of the gray-scale map in Fig. 2 b is launched.

Fig. 4 is the structured flowchart of the image acquiring device of the high spectrum high spatial resolution of the embodiment of the present invention.

Embodiment

Describe embodiments of the invention below in detail, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has the element of identical or similar functions from start to finish.Be exemplary below by the embodiment being described with reference to the drawings, be intended to for explaining the present invention, and can not be interpreted as limitation of the present invention.

In the present invention, unless otherwise clearly defined and limited, the terms such as term " installation ", " being connected ", " connection ", " fixing " should be interpreted broadly, and for example, can be to be fixedly connected with, and can be also to removably connect, or integral; Can be mechanical connection, can be also electrical connection; Can be to be directly connected, also can indirectly be connected by intermediary, can be the connection of two element internals or the interaction relationship of two elements.For the ordinary skill in the art, can understand as the case may be above-mentioned term concrete meaning in the present invention.

Any process of otherwise describing in process flow diagram or at this or method are described and can be understood to, represent to comprise that one or more is for realizing module, fragment or the part of code of executable instruction of step of specific logical function or process, and the scope of the preferred embodiment of the present invention comprises other realization, wherein can be not according to order shown or that discuss, comprise according to related function by the mode of basic while or by contrary order, carry out function, this should be understood by embodiments of the invention person of ordinary skill in the field.

The logic and/or the step that in process flow diagram, represent or otherwise describe at this, for example, can be considered to the sequencing list of the executable instruction for realizing logic function, may be embodied in any computer-readable medium, use for instruction execution system, device or equipment (as computer based system, comprise that the system of processor or other can and carry out the system of instruction from instruction execution system, device or equipment instruction fetch), or use in conjunction with these instruction execution systems, device or equipment.

With regard to this instructions, " computer-readable medium " can be anyly can comprise, device that storage, communication, propagation or transmission procedure use for instruction execution system, device or equipment or in conjunction with these instruction execution systems, device or equipment.The example more specifically (non-exhaustive list) of computer-readable medium comprises following: the electrical connection section (electronic installation) with one or more wirings, portable computer diskette box (magnetic device), random-access memory (ram), ROM (read-only memory) (ROM), the erasable ROM (read-only memory) (EPROM or flash memory) of editing, fiber device, and portable optic disk ROM (read-only memory) (CDROM).In addition, computer-readable medium can be even paper or other the suitable medium that can print described program thereon, because can be for example by paper or other media be carried out to optical scanning, then edit, decipher or process in electronics mode and obtain described program with other suitable methods if desired, be then stored in computer memory.

Should be appreciated that each several part of the present invention can realize with hardware, software, firmware or their combination.In the above-described embodiment, multiple steps or method can realize with being stored in software or the firmware carried out in storer and by suitable instruction execution system.For example, if realized with hardware, the same in another embodiment, can realize by any one in following technology well known in the art or their combination: there is the discrete logic for data-signal being realized to the logic gates of logic function, there is the special IC of suitable combinational logic gate circuit, programmable gate array (PGA), field programmable gate array (FPGA) etc.

Those skilled in the art are appreciated that realizing all or part of step that above-described embodiment method carries is can carry out the hardware that instruction is relevant by program to complete, described program can be stored in a kind of computer-readable recording medium, this program, in the time carrying out, comprises step of embodiment of the method one or a combination set of.

In addition, the each functional unit in each embodiment of the present invention can be integrated in a processing module, can be also that the independent physics of unit exists, and also can be integrated in a module two or more unit.Above-mentioned integrated module both can adopt the form of hardware to realize, and also can adopt the form of software function module to realize.If described integrated module realizes and during as production marketing independently or use, also can be stored in a computer read/write memory medium using the form of software function module.The above-mentioned storage medium of mentioning can be ROM (read-only memory), disk or CD etc.

In the description of this instructions, the description of reference term " embodiment ", " some embodiment ", " example ", " concrete example " or " some examples " etc. means to be contained at least one embodiment of the present invention or example in conjunction with specific features, structure, material or the feature of this embodiment or example description.In this manual, to the schematic statement of above-mentioned term not must for be identical embodiment or example.And, specific features, structure, material or the feature of description can one or more embodiment in office or example in suitable mode combination.In addition, those skilled in the art can carry out combination and combination by the different embodiment that describe in this instructions or example.

The image acquiring method of the high spectrum high spatial resolution of first aspect present invention embodiment as shown in Figure 1, can comprise the following steps:

A. the RGB figure of high spatial resolution and the gray-scale map of high spectral resolution that obtain Same Scene, wherein RGB figure is region-wide sampled images, gray-scale map is the part sampled images through being uniformly distributed rectangle hole mask.

Particularly, can obtain with traditional triple channel color camera the RGB figure of high spatial resolution, as shown in Figure 2 a.Meanwhile, obtain the gray-scale map of high spectral resolution in conjunction with the mask that is evenly distributed with rectangle hole with gray scale camera, as shown in Figure 2 b.Each spatial point in actual scene can find this brightness value under R, G, B triple channel from RGB figure.But in actual scene, only there is segment space point can from gray-scale map, find the monochrome information of the full spectrum of this point.

B. RGB figure alignd with gray-scale map and superposeed, obtaining synthetic image.

Because the locus of triple channel color camera and gray scale camera is not quite identical, so RGB figure need to be alignd with gray-scale map.The mode of alignment can adopt manual demarcation angle point, then obtains the corresponding relation of the pixel on pixel and the gray-scale map on RGB figure according to the systematicness automatic seeking point of sampled point.Can be by partial stack total with gray-scale map RGB figure after alignment, the new images after being alignd is called synthetic image.Now, the one part of pixel point on synthetic image is known brightness value under R, G, B triple channel both, again known in full spectrum the brightness value of different wave length λ.Another part pixel on synthetic image is known brightness value under R, G, B triple channel only, but the brightness value of the unknown different wave length λ in full spectrum.

C. for pixel on synthetic image, have sample information in gray-scale map, directly from gray-scale map, obtain the spectral information of pixel.

Particularly, the pixel on gray-scale map is carried out to spectral evolution, can obtain the brightness value corresponding to light of different wave length.Common spectral evolution can be to assist expansion by corner angle, and this does not repeat for it be known to those skilled in the art that herein.For example, Fig. 3 shows the spectral evolution result of the bright spot (single sampled point) in Fig. 2 b.

D. for the pixel of sample information disappearance on synthetic image, in gray-scale map, the spectral information of the spectrum propagation algorithm calculating pixel point of application based on bilateral filtering.

Particularly, the spectral information of arbitrary target pixel points can obtain according to the spectrum propagation algorithm formula based on bilateral filtering (spatial similarity and color similarity).

First, the object pixel point coordinate that need to ask for spectral information on note synthetic image is (i, j), with (i, j) for the center of circle, paint circle take preset reference distance R as radius, note border circular areas is Ψ, detects and in Ψ, has the reference image of at least one known spectra information vegetarian refreshments k.

Secondly,, according to the spectrum propagation algorithm thought of bilateral filtering, the formula that proposition is calculated target pixel points (i, j) by the spectral information of the reference image vegetarian refreshments k in Ψ is as follows:

${\mathrm{ms}}_{\mathrm{ij}}=\underset{c\∈R,G,B}{\mathrm{\Σ}}\frac{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}{G}_{{\mathrm{\σ}}_r}\left(d_k^{\mathrm{RGB}}\right)G_{{\mathrm{\σ}}_s}\left(d_k\right)\·{\mathrm{\ρ}}_k^c\·{\mathrm{ms}}_k\·r_k^c}{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}{G}_{{\mathrm{\σ}}_r}\left(d_k^{\mathrm{RGB}}\right)G_{{\mathrm{\σ}}_s}\left(d_k\right)}---\left(1\right)$

In above formula: c represents certain passage (being that c passage is a passage in R, G, B triple channel).

represent that respectively average is 0, standard deviation is σ _rand σ _sgauss operator (wherein σ _rand σ _sset manually). the RGB difference of target pixel points and reference image vegetarian refreshments k in expression RGB figure, i.e. color distortion.Calculating

in time, is obtained respectively according to R, G, B passage

d _krepresent the theorem in Euclid space distance of target pixel points and reference image vegetarian refreshments k.

represent the brightness ratio relation of pixel (i, j) and k in specific Color Channel, or take red channel R as example

ms _kthe spectral information that represents k point place is a multi-C vector being made up of the response on each specific wavelength at k point place.

represent that the light that wavelength is λ accounts for its ratio in R, G, the three-channel response ratio of B sum in the response ratio of c passage,

Due to Gauss operator the coefficient of Gauss operator in (1) formula

can in calculating, disappear.(1) formula can further be written as:

${\mathrm{ms}}_{\mathrm{ij}}\left(\mathrm{\λ}\right)=\underset{c\∈R,G,B}{\mathrm{\Σ}}\frac{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})\·{\mathrm{\ρ}}_k^c\·{\mathrm{ms}}_k\left(\mathrm{\λ}\right)\·r_{\mathrm{\λ}}^c}{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{x^2}{{\mathrm{\σ}}_s^2})}---\left(2\right)$

By can obtain the spectral information at (i, j) some place to the calculating of (2) formula.Wherein, ms _ij(λ) brightness value of the light that the wavelength at expression target pixel points place is λ, be illustrated in poor at the pixel value of c passage of target pixel points and reference image vegetarian refreshments k in RGB figure, d _krepresent the theorem in Euclid space distance of target pixel points and reference image vegetarian refreshments k, σ _rrepresent the first default Gauss operator standard deviation, σ _srepresent the second default Gauss operator standard deviation,

represent the brightness ratio of target pixel points and reference image vegetarian refreshments k c passage in RGB figure

the brightness value of the light that the wavelength that represents reference image vegetarian refreshments k place is λ,

represent that the light that wavelength is λ accounts for its ratio in R, G, the three-channel response ratio of B sum in the response ratio of c passage.

For the arbitrary target pixel points (i, j) in synthetic image, its computing method are consistent and separate, thereby are suitable for carrying out speed-up computation with GPU as hardware.If this method is applied to Video processing in addition, for different frame of video, owing to having, the locus of sampled point of spectral information is constant, participates in the sampled point k that calculates in formula, and its corresponding theorem in Euclid space is apart from d _k, wavelength is in R, G, the three-channel response ratio of B

all constant.Therefore in pre-service, just calculate in advance, in spectrum communication process, only need to calculate ms _k(λ).

As from the foregoing, the image acquiring method of the high spectrum high spatial resolution of the embodiment of the present invention, after obtaining the synthetic image after RGB figure and gray-scale map alignment, applies the spectrum propagation algorithm based on bilateral filtering to synthetic image, and accelerate with GPU, obtain the spectral information of each pixel.The method has overcome light spectrum image-forming field and has been difficult to obtain high spatial resolution and a high spectral resolution difficult problem simultaneously, and the development of light spectrum image-forming is had to great importance; The propagation algorithm accelerating has been realized real-time, thereby can obtain high spectrum video; Needed equipment is comparatively simple, and cost is little, after integrated, is convenient to carry out commercial production.

The image acquiring device of the high spectrum high spatial resolution of second aspect present invention embodiment as shown in Figure 4, can comprise: RGB figure acquisition module 10, gray-scale map acquisition module 20, alignment laminating module 30, spectral information computing module 40 and output module 50.

RGB figure acquisition module 10 is for obtaining the RGB figure of high spatial resolution, and wherein RGB figure is region-wide sampled images.

Gray-scale map acquisition module 20 is for obtaining the gray-scale map of high spectral resolution, and wherein gray-scale map is the part sampled images through being uniformly distributed rectangle hole mask.

Alignment laminating module 30 is connected with gray-scale map acquisition module 20 with RGB figure acquisition module 10 respectively, for RGB figure is alignd with gray-scale map and superposeed, obtains synthetic image.

Spectral information computing module 40 is connected with the laminating module 30 that aligns, for the pixel for sample information disappearance on synthetic image, gray-scale map, and the spectral information of the spectrum propagation algorithm calculating pixel point of application based on bilateral filtering.

In one embodiment of the invention, spectral information computing module 40 specifically comprises: reference zone defines submodule and calculating sub module.

Reference zone defines the reference zone that submodule is (i, j) for the object pixel point coordinate that need to ask for spectral information defining on synthetic image.Particularly, reference zone defines submodule for take (i, j) as the center of circle, paint circle take preset reference distance R as radius, and note border circular areas is Ψ, detects and in Ψ, has the reference image of at least one known spectra information vegetarian refreshments k.

Calculating sub module is for calculating the spectral information of target pixel points according to following formula.

${\mathrm{ms}}_{\mathrm{ij}}\left(\mathrm{\λ}\right)=\underset{c\∈R,G,B}{\mathrm{\Σ}}\frac{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})\·{\mathrm{\ρ}}_k^c\·{\mathrm{ms}}_k^c\left(\mathrm{\λ}\right)\·r_{\mathrm{\λ}}^c}{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})}$

Wherein, ms _ij(λ) brightness value of the light that the wavelength at expression target pixel points place is λ,

be illustrated in poor at the pixel value of c passage of target pixel points and reference image vegetarian refreshments k in RGB figure, d _krepresent the theorem in Euclid space distance of target pixel points and reference image vegetarian refreshments k, σ _rrepresent the first default Gauss operator standard deviation, σ _srepresent the second default Gauss operator standard deviation,

represent the brightness ratio of target pixel points and reference image vegetarian refreshments k c passage in RGB figure

the brightness value of the light that the wavelength that represents reference image vegetarian refreshments k place is λ,

represent that the light that wavelength is λ accounts for its ratio in R, G, the three-channel response ratio of B sum in the response ratio of c passage.

Preferably, in this calculating sub module, can adopt GPU is the spectral information that hardware calculates target pixel points.

Output module 50 is connected with spectral information computing module 40 with the laminating module 30 that aligns respectively.For pixel on synthetic image, have sample information in gray-scale map, output module 50 is from spectral information the output of alignment laminating module 30 read pixel points.Wherein reading spectral information specifically comprises: the pixel on gray-scale map is carried out to spectral evolution, read the brightness value corresponding to light of different wave length.For the pixel of sample information disappearance on synthetic image, in gray-scale map, output module 50 is from the spectral information result of calculation of spectral information computing module 40 read pixel points and export.

As from the foregoing, the image acquiring device of the high spectrum high spatial resolution of the embodiment of the present invention, after obtaining the synthetic image after RGB figure and gray-scale map alignment, applies the spectrum propagation algorithm based on bilateral filtering to synthetic image, and accelerate with GPU, obtain the spectral information of each pixel.This device has overcome light spectrum image-forming field and has been difficult to obtain high spatial resolution and a high spectral resolution difficult problem simultaneously, and the development of light spectrum image-forming is had to great importance; The propagation algorithm accelerating has been realized real-time, thereby can obtain high spectrum video; Needed equipment is comparatively simple, and cost is little, after integrated, is convenient to carry out commercial production.

Although illustrated and described embodiments of the invention above, be understandable that, above-described embodiment is exemplary, can not be interpreted as limitation of the present invention, and those of ordinary skill in the art can change above-described embodiment within the scope of the invention, modification, replacement and modification.

Claims (8)

1. an image acquiring method for high spectrum high spatial resolution, is characterized in that, comprises the following steps:

The RGB figure of high spatial resolution and the gray-scale map of high spectral resolution that obtain Same Scene, wherein said RGB figure is region-wide sampled images, described gray-scale map is the part sampled images through being uniformly distributed rectangle hole mask;

Described RGB figure is alignd and superposeed with described gray-scale map, obtain synthetic image;

For pixel on described synthetic image, have sample information in described gray-scale map, directly from described gray-scale map, obtain the spectral information of described pixel;

For the pixel of sample information disappearance on described synthetic image, in described gray-scale map, the spectrum propagation algorithm of application based on bilateral filtering calculates the spectral information of described pixel.

2. image acquiring method according to claim 1, is characterized in that, described for pixel on described synthetic image, have sample information in described gray-scale map, and the spectral information that directly obtains described pixel from described gray-scale map specifically comprises:

Described pixel on described gray-scale map is carried out to spectral evolution, obtain the brightness value corresponding to light of different wave length.

3. image acquiring method according to claim 1 and 2, it is characterized in that, the described pixel for sample information disappearance on described synthetic image, in described gray-scale map, the spectral information that the spectrum propagation algorithm of application based on bilateral filtering calculates described pixel specifically comprises:

Remember that the object pixel point coordinate that need to ask for spectral information on described synthetic image is (i, j), take (i, j) as the center of circle, paint circle take preset reference distance R as radius, note border circular areas is Ψ, detects and in Ψ, has the reference image of at least one known spectra information vegetarian refreshments k;

Calculate the spectral information of described target pixel points according to following formula,

${\mathrm{ms}}_{\mathrm{ij}}\left(\mathrm{\λ}\right)=\underset{c\∈R,G,B}{\mathrm{\Σ}}\frac{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})\·{\mathrm{\ρ}}_k^c\·{\mathrm{ms}}_k^c\left(\mathrm{\λ}\right)\·r_{\mathrm{\λ}}^c}{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})}$

Wherein, ms _ij(λ) brightness value of the light that the wavelength at expression target pixel points place is λ,

be illustrated in poor at the pixel value of c passage of target pixel points described in RGB figure and reference image vegetarian refreshments k, d _krepresent the theorem in Euclid space distance of described target pixel points and reference image vegetarian refreshments k, σ _rrepresent the first default Gauss operator standard deviation, σ _srepresent the second default Gauss operator standard deviation,

represent the brightness ratio of described target pixel points and reference image vegetarian refreshments k c passage in RGB figure

the brightness value of the light that the wavelength that represents reference image vegetarian refreshments k place is λ,

represent that the light that wavelength is λ accounts for its ratio in R, G, the three-channel response ratio of B sum in the response ratio of c passage.

4. according to the image acquiring method of the high spectrum high spatial resolution described in claim 1-3, it is characterized in that, calculate the spectral information of target pixel points take GPU as hardware.

5. an image acquiring device for high spectrum high spatial resolution, is characterized in that, comprising:

RGB figure acquisition module, for obtaining the RGB figure of high spatial resolution, wherein said RGB figure is region-wide sampled images;

Gray-scale map acquisition module, for obtaining the gray-scale map of high spectral resolution, wherein said gray-scale map is the part sampled images through being uniformly distributed rectangle hole mask;

Alignment laminating module, described alignment laminating module is connected with gray-scale map acquisition module with described RGB figure acquisition module respectively, for described RGB figure is alignd and superposeed with described gray-scale map, obtains synthetic image;

Spectral information computing module, described spectral information computing module is connected with the described laminating module that aligns, for the pixel for sample information disappearance on described synthetic image, described gray-scale map, the spectrum propagation algorithm of application based on bilateral filtering calculates the spectral information of described pixel;

Output module, described output module is connected with spectral information computing module with the described laminating module that aligns respectively, for pixel on described synthetic image, have sample information in described gray-scale map, described load module reads spectral information the output of described pixel from described alignment laminating module, for the pixel of sample information disappearance on described synthetic image, in described gray-scale map, described load module reads spectral information result of calculation the output of described pixel from described spectral information computing module.

6. the image acquiring device of high spectrum high spatial resolution according to claim 5, it is characterized in that, described for pixel on described synthetic image, have sample information in described gray-scale map, the spectral information that directly reads described pixel from described alignment laminating module specifically comprises:

Described pixel on described gray-scale map is carried out to spectral evolution, read the brightness value corresponding to light of different wave length.

7. according to the image acquiring device of the high spectrum high spatial resolution described in claim 5 or 6, it is characterized in that, described spectral information computing module specifically comprises:

Reference zone defines submodule, be (i for the object pixel point coordinate that need to ask for spectral information defining on described synthetic image, j) reference zone, with (i, j) for the center of circle, paint circle take preset reference distance R as radius, note border circular areas is Ψ, detects and in Ψ, has the reference image of at least one known spectra information vegetarian refreshments k;

Calculating sub module, described calculating sub module is used for calculating according to following formula the spectral information of described target pixel points,

${\mathrm{ms}}_{\mathrm{ij}}\left(\mathrm{\λ}\right)=\underset{c\∈R,G,B}{\mathrm{\Σ}}\frac{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})\·{\mathrm{\ρ}}_k^c\·{\mathrm{ms}}_k^c\left(\mathrm{\λ}\right)\·r_{\mathrm{\λ}}^c}{{\mathrm{\Σ}}_{k\∈\mathrm{\Ψ}}\exp (-\frac{{\left(d_k^c\right)}^2}{{\mathrm{\σ}}_r^2})\exp (-\frac{d_k^2}{{\mathrm{\σ}}_s^2})}$

Wherein, ms _ij(λ) brightness value of the light that the wavelength at expression target pixel points place is λ,

be illustrated in poor at the pixel value of c passage of target pixel points described in RGB figure and reference image vegetarian refreshments k, d _krepresent the theorem in Euclid space distance of described target pixel points and reference image vegetarian refreshments k, σ _rrepresent the first default Gauss operator standard deviation, σ _srepresent the second default Gauss operator standard deviation,

represent the brightness ratio of described target pixel points and reference image vegetarian refreshments k c passage in RGB figure

the brightness value of the light that the wavelength that represents reference image vegetarian refreshments k place is λ,

represent that the light that wavelength is λ accounts for its ratio in R, G, the three-channel response ratio of B sum in the response ratio of c passage.

8. according to the image acquiring device of the high spectrum high spatial resolution described in claim 5-7, it is characterized in that, in described calculating sub module, calculate the spectral information of target pixel points take GPU as hardware.

Images