CN103900696B - The method of polarization imaging under water of a kind of imitative squill vision polarization antagonism perception - Google Patents

Abstract

The present invention proposes the method for polarization imaging under water of a kind of imitative squill vision polarization antagonism perception, the inventive method simulation squill compound eye, gather antagonism to polarization signal, set up polarization antagonism model, the tuning factor of design antagonism model, the polarization sensitivity of regulation and control all directions, obtain polarization parameter for antagonist, and then obtain degree of polarization parameter and synthesis light intensity parameter, complete the conversion of pixel gray scale from light intensity matrix to imaging surface finally by opto-electronic conversion model, obtain each parametric image.The inventive method enhances the amplitude of polarization response signal, reduces the impact that uneven illumination causes, and improves the contrast between object and background in polarization parameter image, improves the precision of underwater target detection.

Description

The method of polarization imaging under water of a kind of imitative squill vision polarization antagonism perception

Technical field

The present invention relates to Underwater Imaging technical field, specifically refer to the method for polarization imaging under water of a kind of imitative squill vision polarization antagonism perception.

Background technology

Under water in environment due to water body to the absorption of light and scattering process larger, luminous energy decays rapidly in water, under water during the degree of depth of 40 meters, green-yellow light is only had to exist in natural light, imaging polarization measuring technique extends to 7 degree of freedom (light intensity, spectrum, space, degree of polarization, polarization azimuth and polarization ellipticity) quantity of information from three-dimensional (light intensity, spectrum and space), utilizes polarization imaging technology can obtain the detailed polarization data information of each polarization parameter gray-scale map and each observation station intuitively under water simultaneously.But owing to being subject to polarizer in the impact of the factors such as different-waveband transmittance difference, printing opacity heterogeneity, polarization image is caused to there is error in measurement and noise, reduce the accuracy of imaging data, finally cause the traditional Stokes polarization parameter image calculated to there is comparatively big error and noise.In fact, in the underwater environment of low-light (level), strong scattering, a lot of biology all has polarization perception, and they are in perception and utilize tool in polarization information to have great advantage, and biological this vision mechanism is to we providing a referential technology path.

Sea life squill has eye structure the most complicated in the world, linearly polarized light can not only be identified, and can also perception circularly polarized light, they can in low-light (level) by its visual dominance, the prey of various camouflage (color is close with water body color) is captured exactly in strong scattering medium, utilize the transport property of polarized light in scattering medium simultaneously, obtain the depth information between target exactly, and accurately locate, squill is of a great variety, about there is kind more than 400 in the whole world, in literary composition, data are mainly derived from the tooth Squillidae class in hallux toe mantis shrimp Superfamily, they often perch in darker waters, in order to the object of surrounding can be seen clearly, the eyes of squill are evolved become increasingly complex, the strongest to yellowish green polarisation of light perception.Squill compound eye is divided into three regions: central zone, back hemisphere and belly hemisphere, wherein, back and belly hemisphere ommatidium perception linearly polarized light, central zone the 5th and the 6th row ommatidium perception circularly polarized light, the ommatidium of squill is made up of 8 photosensory cells, is respectively R1-R7 and R8.Be divided into three groups according to Microvillares orientation, first group is R1, R4 and R5 photosensory cell, and second group is R2, R3, R6 and R7 photosensory cell, and the 3rd group is R8 photosensory cell.In the photosensory cell of first two groups, Microvillares orientation is in organizing plumbness between interior parallel group, formation can the photosensitive passage of perception pair of orthogonal polarized light, signal forms one group of antagonism, and they are by strengthening certain signal and suppressing the antagonism compute mode of another signal to produce the final output of new response signal as photosensitive passage.When observing same point in space, back and belly hemisphere complete the perception of 30 ° of directions, 120 ° of directions, 75 ° of directions and 165 ° of direction linearly polarized lights by R1-R7 photosensory cell, form four groups of antagonism signals.Central zone the 5th and the 6th row ommatidium perception is left-handed forms two groups of antagonism polarization signals with right-circularly polarized light, and R8 photosensory cell attached at yellowish green light wave be that 1/4 λ wave plate converts the circularly polarized light perceived to linearly polarized light, the direction of linearly polarized light is respectively-45 ° and 45 °.

By imitative squill visual perception mechanism, completing collection and the calculating of liner polarization and circular polarization information simultaneously, more effective polarization parameter for characterizing target polarization information and computing method thereof can being worked out.

Summary of the invention

Technical matters to be solved by this invention is to overcome the deficiencies in the prior art, proposes the method for polarization imaging under water of a kind of imitative squill vision polarization antagonism perception.The inventive method comprises the collection of the antagonism of the polarization under water signal of imitative squill vision polarization opposition perception mechanism and the polarization parameter of opposition under water, degree of polarization parameter, the synthesis light intensity calculation method of parameters of imitative squill vision polarization opposition perception mechanism.

In order to solve the problems of the technologies described above, the technical solution adopted in the present invention is: the method for polarization imaging under water of a kind of imitative squill vision polarization antagonism perception, comprises the steps:

Steps A, by imitative squill vision polarization opposition perception mechanism, gathering six polarization signals, is the linear polarization light intensity I (30 °) of 30 °, 120 °, 75 °, 165 ° four directions, I (120 °), I (75 °), I (165 °) and circular polarization light intensity I that is left-handed, dextrorotation both direction respectively _l, I _r;

Step B, judges whether to collect six polarization signals, if so, performs step e; Otherwise perform step C;

Step C, judges whether the polarization signal gathered contains three linear polarization light intensity and a circular polarization light intensity, if so, performs step D; Otherwise, return execution steps A;

Step D, according to three the linear polarization light intensity gathered and a circular polarization light intensity, by calculating, the linear polarization light intensity do not collected respectively and a circular polarization light intensity;

Step e, forms six groups of polarization antagonism signals;

Every two polarization signals are formed two groups of polarization antagonism signals, comprises four groups of linear polarization antagonism signals and two groups of circular polarization antagonism signals; Described four groups of linear polarization antagonism signals are respectively: 30 °/120 °, 120 °/30 °, 75 °/165 ° and 165 °/75 °; Described two groups of circular polarization antagonism signals are respectively: left-handed/dextrorotation, dextrorotation/left-handed;

Step F, obtains polarization parameter for antagonist;

Six groups of polarization antagonism signals input six polarization antagonism passages respectively, there is in each passage polarization antagonism mathematical model, and tuning factor regulation and control are set in a model to the enhancing of antagonism polarization signal and suppression, six polarization antagonism passages export six polarization parameter for antagonist; Be respectively 30 ° of direction polarization parameter for antagonist S _h, 120 ° of direction polarization parameter for antagonist S _v, left-hand polarization parameter for antagonist S _l, dextropolarization parameter for antagonist S _r, 75 ° of direction polarization parameter for antagonist S _dwith 165 ° of direction polarization parameter for antagonist S _-d;

The mathematic(al) representation of described polarization antagonism mathematical model is:

$S_{{\mathrm{\θ}}_1-{\mathrm{\θ}}_2}=k_{{\mathrm{\θ}}_1}*I\left({\mathrm{\θ}}_1\right)-k_{{\mathrm{\θ}}_2}*I\left({\mathrm{\θ}}_2\right)$

I (θ ₁) and I (θ ₂) represent the polarized light light intensity that incident ray exports after different polarization sheet;

θ ₁and θ ₂represent the angle between the direction of vibration of light and polaroid primary optical axis, meet θ ₁+ θ ₂=90 °, namely ensure I (θ ₁) and I (θ ₂) be one group of polarization antagonism signal;

with for tuning factor, $k_{{\mathrm{\&theta;}}_1}\&GreaterEqual;\mathrm{1,0}<k_{{\mathrm{\&theta;}}_2}\&le;1;$

represent the response signal that polarization antagonism model exports;

Step G, based on six the polarization parameter for antagonist obtained, calculates degree of polarization parameter DOP and synthesis light intensity parameter I;

Step H, utilizes opto-electronic conversion model to convert parameter matrix to pixel gray matrix, obtains S _h, S _v, S _l, S _r, S _d, S _-d, DOP and I eight parametric images.

In step D, a described linear polarization light intensity not collected and a circular polarization light intensity, its concrete grammar is: the principle equaling two orthogonal directions polarized light light intensity sums of being decomposed by it according to total light intensity of random polarization state, and two mutually perpendicular linear polarization light intensity sums deduct a linearly polarized light light intensity and obtain another linearly polarized light light intensity; Two mutually perpendicular linear polarization light intensity sums deduct a circularly polarized light light intensity and obtain another circularly polarized light light intensity.

The invention has the beneficial effects as follows: the present invention proposes the method for polarization imaging under water of a kind of imitative squill vision polarization antagonism perception, the inventive method simulation squill compound eye, gather antagonism to polarization signal, set up polarization antagonism model, the tuning factor of design antagonism model, the polarization sensitivity of regulation and control all directions, obtain polarization parameter for antagonist, and then obtain degree of polarization parameter and synthesis light intensity parameter, complete the conversion of pixel gray scale from light intensity matrix to imaging surface finally by opto-electronic conversion model, obtain each parametric image.The inventive method enhances the amplitude of polarization response signal, reduces the impact that uneven illumination causes, and improves the contrast between object and background in polarization parameter image, improves the precision of underwater target detection.

Accompanying drawing explanation

Fig. 1 is the present invention's process flow diagram of the invention process.

Fig. 2 is imitative squill vision polarization antagonism model of the present invention.

Fig. 3 is antagonism operational mathematics model of the present invention.

Embodiment

Below in conjunction with accompanying drawing, the method for polarization imaging under water of a kind of imitative squill vision polarization antagonism perception that the present invention proposes is described in detail:

Fig. 1 is the process flow diagram that the method for polarization imaging under water of imitative squill vision polarization antagonism perception mechanism is implemented, implementation process antagonism

Mainly comprise the steps:

Steps A, by imitative squill vision polarization opposition perception mechanism, gathering six polarization signals, is the linear polarization light intensity I (30 °) of 30 °, 120 °, 75 °, 165 ° four directions, I (120 °), I (75 °), I (165 °) and left-handed, dextrorotation both direction circular polarization light intensity I respectively _l, I _r;

Step B, judges whether to collect six polarization signals, if so, performs step e, otherwise performs step C;

Step C, judges whether at least comprise three different linear polarization light intensity and a circular polarization light intensity in the polarization signal gathered, if so, performs step D, otherwise, return execution steps A;

Step D, equal the principle of two orthogonal directions polarized light light intensity sums of being decomposed by it according to total light intensity of random polarization state, deduct another linearly polarized light light intensity by two mutually perpendicular linear polarization light intensity sums and can calculate the failed linearly polarized light light intensity of collection; Two mutually perpendicular linear polarization light intensity sums deduct a circularly polarized light light intensity and can calculate the failed circularly polarized light light intensity of another collection;

Step e, every two polarization signals are formed two groups of polarization antagonism signals, four groups of linear polarization antagonism signals and two groups of circular polarization antagonism signals can be formed, four groups of linear polarization antagonism signals are respectively: 30 °/120 °, 120 °/30 °, 75 °/165 ° and 165 °/75 °, and two groups of circular polarization antagonism signals are respectively: left-handed/dextrorotation, dextrorotation/left-handed; Six groups of polarization antagonism signals input six polarization antagonism passages respectively, there is in each passage polarization antagonism mathematical model, and tuning factor regulation and control are set in a model to the enhancing of antagonism polarization signal and suppression, six polarization antagonism passages export and obtain six polarization parameter for antagonist, are respectively 30 ° of direction polarization parameter for antagonist S _h, 120 ° of direction polarization parameter for antagonist S _v, left-hand polarization parameter for antagonist S _l, dextropolarization parameter for antagonist S _r, 75 ° of direction polarization parameter for antagonist S _dwith 165 ° of direction polarization parameter for antagonist S _-d; ;

Step F, based on six the polarization parameter for antagonist obtained, according to Malus law, calculates degree of polarization parameter DOP and synthesis light intensity parameter I;

Step G, utilizes opto-electronic conversion model to convert parameter matrix to pixel gray matrix, obtains S _h, S _v, S _l, S _r, S _d, S _-dand DOP and I_ eight parametric images.The collection of polarization antagonism signal is under water 30 ° of directions and 120 ° of direction signals of simulation back hemisphere ommatidium perception respectively, design employing two groups of linear polarization antagonism signals are respectively: 30 °/120 °, 120 °/30 °, simulation central zone the 5th and the 6th row ommatidium photosensory cell perception left circularly polarized light and right-circularly polarized light, design employing two groups of circular polarization antagonism signals are respectively: left-handed/dextrorotation, dextrorotation/left-handed, 75 ° of directions of simulation belly hemisphere ommatidium perception and 165 ° of direction signals, design employing two groups of linear polarization antagonism signals are respectively: 75 °/165 ° and 165 °/75 °.

Fig. 2 is imitative squill vision polarization antagonism model of the present invention.Its simulation squill polarization antagonism perception mechanism, two orthogonal polarization signals form one group of antagonism and are input in corresponding antagonism passage, design six groups of antagonism signals of employing respectively: 30 °/120 °, 120 °/30 °, 75 °/165 °, 165 °/75 ° and left-handed/dextrorotation, dextrorotation/left-handed, the antagonism computing that six groups of antagonism signals linearly subtract each other in polarization antagonism model, produce the final output of new response signal as antagonism passage, six polarization parameter for antagonist according to exporting calculate degree of polarization parameter, synthesis light intensity parameter and other polarizations characterize parameter signal.

Fig. 3 is the mathematical model that in Fig. 2, one group of antagonism signal produces new response signal by a kind of antagonism compute mode of linearly subtracting each other.By the opponent process of the mutually orthogonal polarization signal of this linear subtractive model simulation squill process, input is one group of polarization antagonism signal, and output is the difference between antagonism signal, and is provided with tuning factor regulation and control in a model to the enhancing of antagonism signal and suppression.Its mathematic(al) representation is:

$S_{{\mathrm{\&theta;}}_1-{\mathrm{\&theta;}}_2}=k_{{\mathrm{\&theta;}}_1}*I\left({\mathrm{\&theta;}}_1\right)-k_{{\mathrm{\&theta;}}_2}*I\left({\mathrm{\&theta;}}_2\right)---\left(1\right)$

In formula, I (θ ₁) and I (θ ₂) represent the polarized light light intensity that incident ray exports after different polarization sheet; θ ₁and θ ₂represent the angle between the direction of vibration of light and polaroid primary optical axis, and meet θ ₁+ θ ₂=90 °, this condition ensures that these two groups of signals are one group of antagonism signal; with for tuning factor, because the polarization sensitivity of different directions is different, therefore the enhancer or inhibitor that tuning factor controls often to organize antagonism signal is set, adopt and based on the machine learning mode of template, each tuning factor is trained respectively, first be determined by experiment interval and the value interval of each tuning factor, calculate the information entropy of corresponding polarization parameter for antagonist image and obtained their optimum solution by Gauss curve fitting curve, the value of tuning factor will determine the output after antagonism computing.

After the input of polarization antagonism model is through polaroid, light intensity is respectively I (30 °), I (120 °), I ' (30 °), I ' (120 °), I (75 °), I (165 °), I ' (75 °), I ' (165 °), I _l, I ' _l, I _rwith I ' _rpolarized light, further, I (30 °)=I ' (30 °), I (120 °)=I ' (120 °), I (75 °)=I ' (75 °), I (165 °)=I ' (165 °), I _l=I ' _land I _r=I ' _r, what draw polarization antagonism sensor model often organizes the output of antagonism signal after antagonism computing, is expressed as:

S _h=k ₁*I(30°)-k ₂*I(120°) （2）

S _v=k ₃*I′(120°)-k ₄*I′(30°) （3）=k ₃*I(120°)-k ₄*I(30°)

S _l=k ₅*I _l-k ₆*I _r（4）

S _r=k ₇*I′ _r-k ₈*I′ _l（5）=k ₇*I _r-k ₈*I _l

S _d=k ₉*I(75°)-k ₁₀*I(165°) （6）

S _-d=k ₁₁*I′(165°)-k ₁₂*I′(75°) （7）=k ₁₁*I(165°)-k ₁₂*I(75°)

In formula, S _h, S _v, S _l, S _r, S _dand S _-dfor polarization parameter for antagonist, k ₁, k ₂..., k ₁₂for tuning factor, represent the enhancer or inhibitor to different directions polarized light range value.Just the polarization parameter for antagonist image characterizing target various information can be calculated by the polarized light light intensity of all directions gathered.If arbitrary linear polarization collecting unit and arbitrary circular polarization collecting unit lost efficacy, implement to not affect overall formation method, consider redundancy design method, total light intensity due to random polarization state equals the light intensity sum of two the orthogonal directions polarized lights decomposed by it, has relational expression I=I (30 °)+I (120 °)=I (75 °)+I (165 °)=I _l+ I _r, polarization parameter for antagonist can be expressed as:

S _h=k ₁*I(30°)-k ₂*I(120°) （8）

S _v=k ₃*I(120°)-k ₄*I(30°) （9）

S _l=k ₅*I _l-k ₆*I _r（10）=k ₅*(I(30°)+I(120°)-I _r)-k ₆*I _r

S _r=k ₇*I′ _r-k ₈*I′ _l（11）=k ₇*I _r-k ₈*(I(30°)+I(120°)-I _r)

S _d=k ₉*I(75°)-k ₁₀*I(165°) （12）=(k ₉+k ₁₀)*I(75°)-k ₁₀*(I(30°)+I(120°))

S _-d=k ₁₁*I′(165°)-k ₁₂*I′(75°) （13）=k ₁₁*(I(30°)+I(120°))-(k ₁₁+k ₁₂)*I(75°)

Therefore, I (30 °), I (120 °), I (75 °), I is utilized _rfour polarization parameters equally can in the hope of six polarization parameter for antagonist.

Degree of polarization parameter is the same with six direction polarization sensitivity parameter can be used to the polarization information characterizing target.Circularly polarized light becomes direction of vibration differs 45 ° linearly polarized light with wave plate optical axis after 1/4 λ wave plate, then partial poolarized light I _onatural light I can be represented as _nwith linearly polarized light I _pthe form of sum, any a branch of light intensity magnitude is that the partial poolarized light of I is through a desirable polaroid, and when between the direction of vibration of light and the primary optical axis of polaroid, angle is θ angle, can obtain according to Malus law through the light intensity I ' of the light penetrated after polaroid be:

$\begin{array}{c}{I}^{\&prime;}\left(\mathrm{\&theta;}\right)=I_n^{\&prime;}\left(\mathrm{\&theta;}\right)+I_p^{\&prime;}\left(\mathrm{\&theta;}\right)\\ =\frac{1}{2}{I}_n+I_p{\cos }^2\mathrm{\&theta;}\\ =\frac{1}{2}I+\frac{1}{2}{I}_p\cos 2\mathrm{\&theta;}\end{array}---\left(14\right)$

Degree of polarization is the ratio of polarized portion light intensity and whole light intensity in light beam, i.e. degree of polarization formula (14) also can be expressed as thus:

$I^{\&prime;}\left(\mathrm{\&theta;}\right)=\frac{1}{2}I(1+\mathrm{DOP}\cos 2\mathrm{\&theta;})---\left(15\right)$

According to formula (15), set up polarization sensitivity parameter S _hand S _vequation is:

Simultaneous formula (16) and formula (17) solve degree of polarization calculation expression is:

$\mathrm{DOP}=\frac{\frac{S_h}{S_v}(k_3-k_4)-(k_1-k_2)}{\frac{\sqrt{3}}{2}[\frac{S_h}{S_v}(k_3+k_4)+(k_1+k_2)]}---\left(18\right)$

The synthesis light intensity I characterizing target intensity signal can be expressed as:

$I=\frac{S_h+S_v}{k_1+k_2}---\left(19\right)$

Thus, six polarization parameter for antagonist S for characterizing target polarization information are obtained herein _h, S _v, S _l, S _r, S _dand S _-dwith degree of polarization parameter type (18) and the synthesis light intensity parameter type (19) characterizing target intensity signal.

Utilize opto-electronic conversion model that the light distribution on imaging surface is converted to pixel gray scale, obtain each parametric image, this conversion is divided into two processes: first light distribution is converted to the radiometric quantities in units of pixel, then the radiometric quantities on pixel is converted to pixel gray scale.Radiation flux on single pixel is the double integral of light distribution on two-dimentional pixel region on pixel, and the Compound trapezoid formula according to double integral asks numerical solution, then Φ _pexpression formula is:

${\mathrm{\&Phi;}}_p=\frac{1}{4}\underset{i=0}{\overset{m}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{\mathrm{ij}}I(i,j)---\left(20\right)$

In formula: pixel has the capable n row of m, the light intensity that I (i, j) is pixel, λ _ijfor the i-th row jth column element of the trapezoidal matrix of coefficients of complexification.

Be approximately linear from pixel radiation flux to the transfer process of gray scale, first analog voltage is

$v=\mathrm{\&eta;\&lambda;}\frac{{\mathrm{\&Phi;}}_p\&CenterDot;T_{\exp }}{\mathrm{hc}}---\left(21\right)$

In formula: T _expfor the time shutter, wavelength X gets 550nm, h and c is respectively Planck constant and the light velocity, and η is the slope that electric light is converted to the curve linear section of voltage, gets 11.5 μ V/e.The analog voltage obtained is changed into intensity profile according to the set amount of grey exponent number, finally complete the display of polarization parameter for antagonist image, degree of polarization image, synthesis plot of light intensity picture under water, for a series of images process such as follow-up target detection provide the original image of high-contrast.

Claims (2)

1. the method for polarization imaging under water of imitative squill vision polarization antagonism perception, is characterized in that, comprise the steps:

Steps A, by imitative squill vision polarization opposition perception mechanism, gathering six polarization signals, is the linear polarization light intensity I (30 °) of 30 °, 120 °, 75 °, 165 ° four directions, I (120 °), I (75 °), I (165 °) and circular polarization light intensity that is left-handed, dextrorotation both direction respectively;

Step B, judges whether to collect six polarization signals, if so, performs step e; Otherwise perform step C;

Step C, judges whether the polarization signal gathered contains three linear polarization light intensity and a circular polarization light intensity, if so, performs step D; Otherwise, return execution steps A;

Step D, according to three the linear polarization light intensity gathered and a circular polarization light intensity, by calculating, the linear polarization light intensity do not collected respectively and a circular polarization light intensity;

Step e, forms six groups of polarization antagonism signals;

Every two polarization signals are formed two groups of polarization antagonism signals, comprises four groups of linear polarization antagonism signals and two groups of circular polarization antagonism signals; Described four groups of linear polarization antagonism signals are respectively: 30 °/120 °, 120 °/30 °, 75 °/165 ° and 165 °/75 °; Described two groups of circular polarization antagonism signals are respectively: left-handed/dextrorotation, dextrorotation/left-handed;

Step F, obtains polarization parameter for antagonist;

Six groups of polarization antagonism signals input six polarization antagonism passages respectively, there is in each passage polarization antagonism mathematical model, and tuning factor regulation and control are set in a model to the enhancing of antagonism polarization signal and suppression, six polarization antagonism passages export six polarization parameter for antagonist; Be respectively 30 ° of direction polarization parameter for antagonist S _h, 120 ° of direction polarization parameter for antagonist S _v, left-hand polarization parameter for antagonist S _l, dextropolarization parameter for antagonist S _r, 75 ° of direction polarization parameter for antagonist S _dwith 165 ° of direction polarization parameter for antagonist S _-d;

The mathematic(al) representation of described polarization antagonism mathematical model is:

$S_{{\mathrm{\&theta;}}_1-{\mathrm{\&theta;}}_2}=k_{{\mathrm{\&theta;}}_1}*I\left({\mathrm{\&theta;}}_1\right)-k_{{\mathrm{\&theta;}}_2}*I\left({\mathrm{\&theta;}}_2\right)$

I (θ ₁) and I (θ ₂) represent the polarized light light intensity that incident ray exports after different polarization sheet;

θ ₁and θ ₂represent the angle between the direction of vibration of light and polaroid primary optical axis, meet θ ₁+ θ ₂=90 °, namely ensure I (θ ₁) and I (θ ₂) be one group of polarization antagonism signal;

with for tuning factor, $k_{{\mathrm{\&theta;}}_1}\&GreaterEqual;1,$ $0<k_{{\mathrm{\&theta;}}_2}\&le;1;$

represent the response signal that polarization antagonism model exports;

Step G, based on six the polarization parameter for antagonist obtained, according to Malus law, calculates degree of polarization parameter DOP and synthesis light intensity parameter I;

Step H, utilizes opto-electronic conversion model to convert parameter matrix to pixel gray matrix, obtains S _h, S _v, S _l, S _r, S _d, S _-d, DOP and I eight parametric images.

2. the method for polarization imaging under water of a kind of imitative squill vision polarization antagonism according to claim 1 perception, it is characterized in that, in step D, a described linear polarization light intensity not collected and a circular polarization light intensity, its concrete grammar is: two mutually perpendicular linear polarization light intensity sums deduct a linearly polarized light light intensity and obtain another linearly polarized light light intensity; Two mutually perpendicular linear polarization light intensity sums deduct a circularly polarized light light intensity and obtain another circularly polarized light light intensity.