KR102676333B1 - Polarization Camera for Acquiring Full Stokes Polarized Image, and Polarized Image Acquisition Method using the Same - Google Patents

Abstract

The present invention is an image sensor, disposed in front of the image sensor, and divided into a plurality of polarization regions with different polarization directions according to the number of light intensity images for each polarization direction required to obtain full Stokes parameters, and the image sensor It includes a polarization filter that filters the incident light according to the polarization direction of the divided polarization region and a phase mask that is disposed to overlap the polarization filter in front of the image sensor and phase-converts the incident light according to a pre-formed phase conversion pattern. Accordingly, a polarization camera that can be manufactured at low cost and can easily obtain full Stokes parameters with a single shooting and an image acquisition method using the same are provided.

Description

Polarization Camera for Acquiring Full Stokes Polarized Image, and Polarized Image Acquisition Method using the Same}

The present invention relates to a polarization camera and a method of acquiring an image using the same. It relates to a polarization camera capable of acquiring full Stokes parameters through a single shooting and a method of acquiring a polarization image using the same.

Polarization imaging is an imaging technique that obtains images of an object in various polarization directions to reveal additional contrast in order to observe polarization characteristics that vary depending on the object's birefringence or surface properties.

Figure 1 shows an example application of polarization imaging.

In Figure 1, (a) is an image acquired using a conventional imaging technique, and (b) is an image acquired using a polarization imaging technique. As shown in Figure 1 (b), the use of polarization imaging techniques is becoming widely used due to the advantage of being able to identify hidden information that is difficult to obtain with the human eye or existing cameras. Polarization imaging is widely used in a variety of machine vision applications such as materials analysis, transparent object inspection, and 3D imaging.

Most existing polarization cameras use one of two techniques: spatial multiplexing of the sensor plane or sequential polarization filtering to acquire polarization images in multiple polarization directions. Spatial multiplexing uses an on-chip pixel-level polarization filter array and uses a polarization image sensor in which the image sensor and polarization filter are manufactured as one piece, enabling snapshot polarization imaging.

Figure 2 shows an example of a conventional polarization image sensor configuration.

As shown in FIG. 2, in the polarization image sensor used in the spatial multiplexing method, the polarization filter 12 must be configured in pixel units and combined with the image sensor 11 to be configured on-chip. At this time, the lens 13 may also be manufactured and combined at the pixel level. In addition, in Figure 2, it is composed of a linear polarization filter that acquires linear polarization in four directions, but in order to acquire circular polarization or elliptical polarization together, a 1/4 phase retardation plate ( Quarter-Wave Plate (hereinafter referred to as QWP) must be additionally configured on-chip. Therefore, the manufacturing process is very complicated and the cost is very high. In addition, since it must be manufactured as a dedicated chip for polarization imaging, there is a limitation that it cannot be used for other purposes.

In contrast, sequential polarization filtering is a method of repeatedly photographing multiple times while replacing or rotating polarizing films with different polarization directions on the front of a typical camera. Sequential polarization filtering has the advantage of being able to acquire polarization images at low cost because it requires only an additional rotating polarizer while using an existing camera. However, the sequential polarization filtering method takes a lot of time because it must be imaged multiple times according to different polarization directions, and object movement must not occur between shooting intervals. This limits its applicability to static and/or very slowly moving objects.

Korea Registered Patent No. 10-1999224 (registered on July 5, 2019)

The purpose of the present invention is to provide a polarization camera capable of acquiring full Stokes parameters in one shot using a lensless camera, and a method for acquiring a polarization image using the same.

Another object of the present invention is to provide a polarization camera capable of obtaining full Stokes parameters at low cost using a phase mask and a polarizing film instead of a lens, and a method for acquiring a polarized image using the same.

A polarization camera according to an embodiment of the present invention for achieving the above object includes an image sensor; It is disposed in front of the image sensor and is divided into a plurality of polarization regions with different polarization directions according to the number of light intensity images for each polarization direction required to obtain the full Stokes parameter, so that the light incident on the image sensor A polarization filter that filters according to the polarization direction of the divided polarization region; and a phase mask disposed in front of the image sensor to overlap the polarization filter and phase-converting incident light according to a pre-formed phase conversion pattern.

The polarization filter includes a polarizing film divided into a plurality of polarization regions each having a designated polarization direction so that polarization according to the polarization regions is incident on the light sensing area of the image sensor. and a phase retardation plate disposed to overlap at least one polarization region among the divided polarization regions of the polarizing film to delay the phase of incident light.

The phase mask may be implemented as a transparent film formed to have different heights on one side at each position according to the phase conversion pattern.

The polarization camera divides the point spread function corresponding to the phase shift pattern of the phase mask into a plurality of partial point spread functions according to the plurality of polarization regions of the polarization filter, and divides the point spread function corresponding to the phase shift pattern of the phase mask into a plurality of partial point spread functions for the raw image acquired from the image sensor. It may further include a processor that reconstructs the plurality of light intensity images by individually applying each partial point spread function.

The processor may obtain the full Stokes parameter by combining the plurality of light intensity images.

The polarizing film is divided into four polarization regions, and the four polarization regions may have polarization directions of 0 degrees, 45 degrees, 90 degrees, and 135 degrees, or polarization directions of 0 degrees, 90 degrees, and two 45 degrees.

The phase retardation plate may be disposed to overlap one of two polarization regions having a 45-degree polarization direction or a polarization region having a 135-degree polarization direction in the polarizing film.

The processor generates three light intensity images (I(0°, 0°), I(45°, A light intensity image (I(45°, 90°)), the four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) that make up the full Stokes parameters are expressed in the equation

It can be obtained according to .

The polarization camera includes an aperture in which the polarization filter and the phase mask are combined on one side and the other, respectively; and a spacer that supports the aperture in which the polarization filter and the phase mask are combined to be spaced apart from the front of the image sensor.

A polarization image acquisition method according to another embodiment of the present invention to achieve the above object is a plurality of polarization regions having different polarization directions according to the number of light intensity images for each polarization direction required to obtain the full Stokes parameter. The image sensor detects the incident light through a polarization filter that filters the separately incident light according to the polarization direction of the divided polarization region, and a phase mask that phase-converts the incident light according to a pre-formed phase conversion pattern. acquiring an image; Divide the point spread function corresponding to the phase shift pattern of the phase mask into a plurality of partial point spread functions according to the plurality of polarization regions of the polarization filter, and apply each of the plurality of partial point spread functions individually to the raw image. reconstructing a plurality of light intensity images; and obtaining the full Stokes parameter by combining the plurality of light intensity images.

Therefore, the polarization camera and the method of acquiring a polarization image using the same according to an embodiment of the present invention enable various polarization images of different polarization directions to be acquired in a single shooting using a phase mask and a polarizing film in a lensless camera. Therefore, it can be manufactured at a low cost, and the full Stokes parameters can be easily obtained with a single shooting, thereby obtaining a polarization degree image, a polarization angle image, and an ellipticity image to explain the polarization state.

Figure 1 shows an example application of polarization imaging.
Figure 2 shows an example of a conventional polarization imaging camera configuration.
Figure 3 is a diagram for explaining the operation of a lens-based imaging method and a phase mask-based lensless imaging method.
Figure 4 is a diagram to explain the principle of a phase mask-based lensless imaging method.
Figure 5 is a diagram for explaining a method for restoring a phase mask-based lensless image.
Figure 6 shows a schematic structure of a polarization camera according to an embodiment of the present invention.
FIG. 7 is a diagram to explain the concept of reconstructing intensity images according to various polarization directions using the polarization camera of FIG. 6.
FIG. 8 is a diagram for explaining a method of acquiring a polarization state image based on the reconstructed intensity image of FIG. 7.

In order to fully understand the present invention, its operational advantages, and the objectives achieved by practicing the present invention, reference should be made to the accompanying drawings illustrating preferred embodiments of the present invention and the contents described in the accompanying drawings.

Hereinafter, the present invention will be described in detail by explaining preferred embodiments of the present invention with reference to the accompanying drawings. However, the present invention may be implemented in many different forms and is not limited to the described embodiments. In order to clearly explain the present invention, parts that are not relevant to the description are omitted, and like reference numerals in the drawings indicate like members.

Throughout the specification, when a part "includes" a certain element, this does not mean excluding other elements, unless specifically stated to the contrary, but rather means that it may further include other elements. In addition, terms such as "... unit", "... unit", "module", and "block" used in the specification refer to a unit that processes at least one function or operation, which is hardware, software, or hardware. and software.

In the polarization camera and polarization image acquisition method for acquiring the full Stokes parameters of this embodiment, the full Stokes parameters can be acquired using a phase mask-based lensless camera along with a polarizing film.

Therefore, here we will first explain the phase mask-based lensless camera.

FIG. 3 is a diagram for explaining the operation of the lens-based imaging method and the phase mask-based lensless imaging method, FIG. 4 is a diagram for explaining the principle of the phase mask-based lensless imaging method, and FIG. 5 is a diagram for explaining the operation of the phase mask-based lensless imaging method. This diagram is intended to explain how to restore a lease image.

In Figure 3, (a) shows a lens-based imaging method, and (b) shows a phase mask-based lensless imaging method.

As shown in (a) of FIG. 3, in the lens-based imaging method, a lens module 21 including at least one lens is disposed in front of the image sensor 22. The lens module 21 allows light incident from a specific location to be imaged at a specific location on the image sensor 22. That is, since the point light source is focused again in the form of a point on the image sensor 22 by the lens module 21, the image sensor 22 can acquire an image 23 of the shape of the object. On the other hand, in the phase mask-based lensless imaging method, as shown in (b), a phase mask 24 is located in front of the image sensor 25 instead of a lens. Here, the phase mask 24 is a transparent material capable of transmitting light and can be implemented using a transparent film, etc., and as shown in (b) of FIG. 3, one side of the phase mask 24 has an irregular shape. It is formed to have different heights for each location according to the phase conversion pattern.

The phase mask 24, which has various patterns of different sizes, heights, and shapes formed on one surface, refracts light with different refractive indices depending on the position where the light is incident, and as shown in (a) of FIG. 4, the incident point Converts the light source to spread out in a pattern similar to a wave pattern. That is, the point light source is acquired as an image spread over the entire area of the image sensor according to the phase shift pattern formed on the phase mask 24.

Here, the Point Spread Function (PSF) 28 is a function representing the distribution of light quantity incident on the image sensor 25 when light emitted from a point light source at a predetermined reference position passes through an optical system (here, a phase mask). As shown in (a) of FIG. 4, a pattern corresponding to the phase shift pattern of the phase mask 24 is obtained in advance.

And when the position of the incident point light source changes, as shown in (b) of FIG. 4, the entire point spread function (PSF) projected on the image sensor 25 changes in a parallel shift form. do. Therefore, the light incident from the point light sources at different positions is converted to diffusion according to the point spread function (PSF) and is overlapped and incident on the image sensor 25, so the image sensor 25 is as shown in (b) of FIG. The point spread function (PSF) is parallel shifted to obtain a superimposed raw image (26).

When using the phase mask 24 in this way, the point light source is diffused and projected onto the entire area of the image sensor 25 according to the phase conversion pattern, so the point light source can be projected only in a partial area of the raw image acquired from the image sensor 25. It can be reorganized to a certain level.

Additionally, when photographing an object having a size other than a point, light emitted from a plurality of point light sources at each location can be viewed as being overlapped and incident on the image sensor 25 through the phase mask 24. Accordingly, a raw image 26 such as (b) of FIG. 3 is obtained by shifting and overlapping multiple point spread functions (PSFs). At this time, since some areas of the raw image 26 also contain components of multiple point light sources in an overlapping form, the image of the object can be reconstructed to a certain level with only some areas of the raw image 26.

Here, in order to convert the raw image 26 acquired by the lensless imaging method into an image of the same form as the lens-based imaging method, a convolution operation is first performed on the raw image 26 using a point spread function (PSF). After extracting the image from each point light source, a deconvolution operation must be performed again based on the point spread function (PSF).

However, the deconvolution operation based on the point spread function (PSF) is very complex and cannot be performed in a conventional manner due to limitations in the area in which the image sensor 25 can detect light. Accordingly, the phase-converted raw image 26 is generally reconstructed into a lens-based image form 29 using an alternating direction method of multipliers (ADMM) algorithm. In this case, an optimization method is used to preset the image conversion model and iteratively update the image conversion model so that the error between the image converted from the image conversion model and the lens-based image is minimized, through the phase mask 24. The acquired raw image 26 can be reconstructed into a lens-based image form 29 as shown in FIG. 5.

Meanwhile, in polarization imaging, the sequential polarization filtering method requires repeated imaging multiple times while replacing or rotating the polarizing film, as described above. This is because, when using a typical lens camera, images of the entire shape of an object cannot be acquired simultaneously with polarizing films in different polarization directions. However, when the phase mask 24 is applied, as described above, the phase mask 24 diffuses and converts a plurality of point light sources at each location according to the phase conversion pattern and projects them on the image sensor, so the image sensor The diffusion function (PSF) is shifted and a superimposed raw image 26 is obtained. In this way, in the case of the raw image 26 in which the point spread function (PSF) is shifted and overlapped, the approximate shape of the object can be obtained from only a partial area.

Therefore, after dividing the entire area of the image sensor 25 that detects the light passing through the phase mask 24 and applying polarization filters in different directions to each divided area, the raw data obtained from each divided area is If images are restored independently, object images for multiple different polarization directions can be obtained in one shot.

Figure 6 shows a schematic structure of a polarization camera according to an embodiment of the present invention.

Referring to FIG. 6, in the polarization camera according to the present embodiment, the image sensor 100 is disposed on a substrate 500, and the image sensor 100 is positioned at a predetermined distance in front of the image sensor 100 (here, the focus distance is 6 mm, for example). Phase masks 300 are disposed spaced apart. As described above, the phase mask 300 has various patterns of different sizes, heights, and shapes formed on one surface so that the incident light is converted into diffusion according to a point spread function. As shown in FIG. 6, the phase mask 300 is coupled to the aperture 410, and the phase mask 300 coupled to the aperture 410 is connected to the image sensor (200) by the spacer 420. 100) may be arranged to be spaced apart from a predetermined distance.

Meanwhile, the film-shaped polarizing filter 200 may also be coupled to the aperture 410 and placed at a predetermined distance from the image sensor 100. Therefore, in the polarization camera of this embodiment, the polarization filter 200 is not configured on-chip with the image sensor 100, but is configured in the form of a separate film and is disposed to be spaced apart from the image sensor 100. Additionally, the polarizing filter 200 may include a polarizing film 210 and a phase retardation plate (Quarter-Wave Plate: QWP) 220.

First, the polarizing film 210 is configured to have a predetermined polarization direction in each of the predetermined number of divided areas of the entire area where light is detected in the image sensor 100. Here, as an example, it is assumed that the polarizing film 210 has four polarization regions 211 to 214, each having a designated polarization direction. However, the polarization regions may be divided into six or eight, etc. And among the four polarization areas 211 to 214, the first polarization area 211 has a polarization direction of 0 degrees, the third polarization area 213 has a polarization direction of 90 degrees, and the second and fourth polarization areas 212 , 214) is assumed to have the same 45 degree polarization direction. Here, among the four polarization regions 211 to 214, the second and fourth polarization regions 212 and 214 have the same 45-degree polarization direction, and the phase retardation plate 220 is used together with the fourth polarization region 214. This is to obtain a circularly polarized image using. That is, the first to third polarization areas 211 to 213 are used to obtain linearly polarized images in different directions, and the fourth polarization area 214 is used to obtain a circularly polarized image. However, the fourth polarization area 214 may have a different polarization direction (for example, 135 degrees).

Meanwhile, the phase retardation plate 220 overlaps at least one polarization region (here, as an example, the fourth polarization region 214) among the plurality of polarization regions 211 to 214 of the polarization film 210 to obtain circularly polarized light. It is placed. The combination of a linear polarization filter and a phase retardation plate 220 is a well-known technique for obtaining circular polarization. The 45 degree linear polarization filter and the phase retardation plate 220 can obtain left-circular polarization, and the 135 degree linear polarization filter and the phase retardation plate 220 can obtain right-circular polarization.

In general, the polarization state is circular polarization, with a Degree of Polarization (DoP) image indicating the degree of polarization and an Angle of Polarization (AoP) image indicating the angle of linear polarization, as shown in Figure 1. It can be expressed as an ellipticity image representing the ellipticity of . And the DoP image, AoP image, and ellipticity image can be obtained from the full Stokes parameter consisting of four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ), and the four Stokes parameters of the full Stokes parameter (S ₀ , S ₁ , S ₂ , S ₃ ) can be obtained from multiple light intensity images (I(θ, ϕ)) according to the linear polarization angle (θ) and phase retardation angle (ϕ). there is.

The four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) are calculated from eight light intensity images (I(0°,0°), I(45°,0°), I(90°,0°). ), I(135°,0°), I(0°,90°), I(45°,90°), I(90°,90°), I(135°,90°)) You can. That is, light intensity images according to four linear polarizations (I(0°,0°), I(45°,0°), I(90°,0°), I(135°,0°)) and 90° Light intensity images (I(0°,0°), I(45°,0°), I(90°,0°), I(135°,0°)) according to circular polarization with additional phase delay reflected. Based on this, four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) can be obtained.

Here, four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) and eight light intensity images (I(0°,0°), I(45°,0°), I(90°,0 °), I(135°,0°), I(0°,90°), I(45°,90°), I(90°,90°), I(135°,90°)) The relationship is as shown in Equation 1.

And Equation 1 can be organized into equations for the four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ), as in Equation 2.

According to Equation 2, eight light intensity images (I(0°,0°), I(45°,0°), I(90°,0°), I(135°,0°), I( Four light intensity images (I(0°,0°), I(45°,90°), I(90°,90°), I(135°,90°)) Four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) are obtained using only I(45°,0°), I(90°,0°), and I(45°,90°)), It can be seen that DoP images, AoP images, and ellipticity images to explain the polarization state can be obtained.

Accordingly, in this embodiment, the polarization filter 200 is provided with four polarization regions 211 to 214 having linear polarization and a phase retardation plate 220 to produce four light intensity images (I(0°, 0°), I(45°,0°), I(90°,0°), and I(45°,90°)) are shown assuming the configuration is possible.

However, if the number of light intensity images according to different polarization directions is 6 or 8, four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) can be obtained with fewer operations, so the above As shown, the polarizing film 210 may be divided into 6 or 8 regions.

If the polarizing film 210 is divided into six areas, it may include two polarization areas in directions of 45 degrees and 135 degrees, and if it is divided into eight areas, it may include polarization areas in directions of 0 degrees, 45 degrees, 90 degrees, and 135 degrees. It may also include two polarization regions in each direction. Accordingly, when the polarizing film 210 is divided into six regions, the phase retardation plate 220 can be arranged to overlap each of the two 45-degree and 135-degree polarization regions, which are divided into eight regions. In this case, a circularly polarized image may be obtained by arranging the two polarization regions to overlap one of each of the 0-degree, 45-degree, 90-degree, and 135-degree directions.

Therefore, in this embodiment, as shown in FIG. 6, the polarization filter 200 is not configured to have different polarization directions at the pixel level of the image sensor 100, but is configured to have different polarization directions in the image sensor 100 according to the polarization direction of the polarization image to be obtained. It is composed by dividing the entire area of 100. Additionally, the phase mask 300 can be easily implemented with polydimethylsiloxane (PDMS), etc. Therefore, the polarization camera of the present invention consists of a general camera without a lens, a polarization filter 200, and a phase mask 300, and can be manufactured at a very low cost.

Meanwhile, the polarization camera of this embodiment is not shown, but the raw image 600 acquired from the image sensor 100 is converted into a plurality of light intensity images (I(0°, 0°), I(45°, 0°), I (90°, 0°), I(45°, 90°)) may further include a processor (not shown). The processor then calculates the full Stokes parameters from multiple light intensity images (I(0°,0°), I(45°,0°), I(90°,0°), I(45°,90°)). (S ₀ , S ₁ , S ₂ , S ₃ ) can be acquired, and in some cases, DoP images, AoP images, and ellipticity images can be obtained from full Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ). You can also obtain it.

FIG. 7 is a diagram to explain the concept of reconstructing intensity images according to various polarization directions using the polarization camera of FIG. 6.

In FIG. 6, the polarization filter 200 is shown as being disposed between the phase mask 300 and the image sensor 100. However, the phase mask 300 is disposed between the polarization filter 200 and the image sensor, as shown in FIG. 7. It may be placed between (100).

Referring to FIG. 7, the light incident from the polarization camera of the present invention is diffusely converted through the phase mask 300 and incident on the image sensor 100. At this time, the light is diffusely converted according to the point spread function (PSF). do. Accordingly, the raw image 600 detected by the image sensor 100 can be reconstructed in the form of an image from a lens camera using a point spread function (PSF). However, in this embodiment, the polarization filter 200 filters the incident light into different polarizations for each divided area, so the raw image 600 detected by the image sensor 100 is also filtered into the divided area of the polarization filter 200. It must be divided and reorganized accordingly.

Accordingly, in this embodiment, the point spread function (PSF) of the phase mask 300 is divided into a plurality of points according to a plurality of divided areas of the polarization filter 200. Here, the point spread function (PSF) of the phase mask 300 is divided into four partial point spread functions (PSF(0°,0°), PSF(45°,0) corresponding to the four distinct areas of the polarization filter 200. °), PSF(90°,0°), PSF(45°,90°)). and distinct first to fourth partial point spread functions (PSF(0°,0°), PSF(45°,0°), PSF(90°,0°), PSF(45°,90°)) respectively. is individually applied to the corresponding region of the raw image 600 to produce four light intensity images (I(0°,0°), I(45°,0°), I(90°,0°), I( 45°,90°)) can be reconstructed. At this time, the four light intensity images (I(0°,0°), I(45°,0°), I(90°,0°), I(45°,90°)) are calculated using the ADMM algorithm, etc. It can be obtained, and since this is a known technique, it will not be described in detail here.

FIG. 8 shows a method of acquiring a polarization state image based on the reconstructed intensity image of FIG. 7 and an example of the obtained polarization state image.

Four light intensity images (I(0°,0°), I(45°,0°), I(90°,0°), I(45°,90°)) are reconstructed from the raw image 600. Once acquired, four light intensity images (I(0°,0°), I(45°,0°), I(90°,0°), I(45°,90°) are obtained according to Equation 2. ) By combining the full Stokes parameters, that is, four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) can be obtained. And when _the four Stokes parameters (S ₀ , S ₁ , S ₂ , _{S 3 ) are} obtained _, the DoP image and Acquire AoP images and elliptically polarized images.

As shown in FIGS. 7 and 8 , in this embodiment, a raw image 600 is acquired using a lensless camera including a phase mask 300 and a polarizing filter 200 divided into regions. And for the acquired raw image 600, a point spread function (PSF) according to the phase mask 300 is divided into a plurality of partial point spread functions (PSF (0°, By reconstructing the lens image using 0°), PSF(45°,0°), PSF(90°,0°), and PSF(45°,90°)), DoP image, AoP image and Full Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) that can acquire elliptically polarized images can be obtained.

The method according to the present invention can be implemented as a computer program stored on a medium for execution on a computer. Here, computer-readable media may be any available media that can be accessed by a computer, and may also include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data, including read-only memory (ROM). It may include dedicated memory), RAM (random access memory), CD (compact disk)-ROM, DVD (digital video disk)-ROM, magnetic tape, floppy disk, optical data storage, etc.

The present invention has been described with reference to the embodiments shown in the drawings, but these are merely exemplary, and those skilled in the art will understand that various modifications and other equivalent embodiments are possible therefrom.

Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the attached claims.

100: image sensor 200: polarizing filter
210: Polarizing film 211 ~ 214: Polarization area
220: phase retardation plate 300: phase mask
410: Aperture 420: Spacer
500: substrate

Claims (16)

image sensor;
It is disposed in front of the image sensor and is divided into a plurality of polarization regions with different polarization directions according to the number of light intensity images for each polarization direction required to obtain the full Stokes parameter, so that the light incident on the image sensor A polarization filter that filters according to the polarization direction of the divided polarization region; and
A phase mask disposed in front of the image sensor to overlap the polarization filter and phase-converting incident light according to a pre-formed phase conversion pattern,
The image sensor acquires a raw image by detecting light incident through the phase mask,
Divide the point spread function corresponding to the phase shift pattern of the phase mask into a plurality of partial point spread functions according to the plurality of polarization regions of the polarization filter, and apply each of the plurality of partial point spread functions individually to the raw image. A polarization camera further comprising a processor configured to reconstruct the plurality of light intensity images and obtain the full Stokes parameter by combining the plurality of light intensity images. The method of claim 1, wherein the polarizing filter is
a polarizing film divided into a plurality of polarization regions each having a designated polarization direction so that polarization according to the polarization regions is incident on the light sensing area of the image sensor; and
A polarization camera including a phase retardation plate disposed to overlap at least one polarization region among the divided polarization regions of the polarization film and delaying the phase of incident light. The method of claim 2, wherein the phase mask is
A polarizing camera implemented as a transparent film formed to have different heights for each position on one side according to the phase conversion pattern. delete delete The method of claim 1, wherein the polarizing film
A polarizing camera divided into four polarization zones, the four polarization zones having 0 degrees, 45 degrees, 90 degrees, and 135 degrees polarization directions, or 0 degrees, 90 degrees, and two 45 degrees polarization directions. The method of claim 6, wherein the phase retardation plate is
A polarization camera disposed to overlap one of two polarization regions having a 45-degree polarization direction or a polarization region having a 135-degree polarization direction in the polarizing film. The method of claim 7, wherein the processor
Three light intensity images (I(0°,0°), I(45°,0°) acquired by light incident on the image sensor through polarization regions having 0, 45, and 90 degree polarization directions. , I (90°, 0°)) and a polarization region having the 45 degree polarization direction and a light intensity image (I (45°, 90°)) obtained by light incident on the image sensor through the phase retardation plate. ), the four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) that make up the full Stokes parameters are expressed in the equation

Polarization camera that acquires according to. The method of claim 1, wherein the polarization camera
an aperture in which the polarizing filter and the phase mask are combined on one side and the other, respectively; and
A polarization camera further comprising a spacer that supports the aperture combined with the polarization filter and the phase mask to be spaced apart from the front of the image sensor. In a method of acquiring a polarized image of a polarization camera including a processor that processes a raw image acquired from an image sensor,
A polarization filter that divides incident light into a plurality of polarization regions with different polarization directions according to the number of light intensity images for each polarization direction required to obtain full Stokes parameters and filters the incident light according to the polarization direction of the divided polarization regions. and obtaining a raw image by the image sensor detecting incident light through a phase mask that phase-converts incident light according to a pre-formed phase conversion pattern;
Divide the point spread function corresponding to the phase shift pattern of the phase mask into a plurality of partial point spread functions according to the plurality of polarization regions of the polarization filter, and apply each of the plurality of partial point spread functions individually to the raw image. reconstructing a plurality of light intensity images; and
A polarization image acquisition method comprising obtaining the full Stokes parameter by combining the plurality of light intensity images. The method of claim 10, wherein acquiring the raw image comprises
A polarization image acquisition method in which polarized light filtered through a plurality of polarization regions of the polarization filter is phase-converted through the phase mask, and the image sensor detects the phase-converted polarization for each region to generate the raw image. The method of claim 10, wherein acquiring the raw image comprises
A polarization image acquisition method in which the image sensor detects polarized light phase-converted through the phase mask and filtered through a plurality of polarization regions of the polarization filter to generate the raw image. The method of claim 10, wherein the phase mask is
A polarized image acquisition method implemented with a transparent film formed to have different heights for each position on one side according to the phase conversion pattern. The method of claim 10, wherein acquiring the raw image comprises
A polarization image acquisition method in which the image sensor acquires the raw image by applying phase-delayed elliptical polarization after filtering 0 degree, 45 degree, and 90 degree linear polarization and 45 degree linear polarization filtered by the polarization filter. The method of claim 14, wherein acquiring the light intensity image comprises
In the raw image, three light intensity images ( Reconstructing I(0°,0°), I(45°,0°), I(90°,0°)); and
Polarization comprising the step of reconstructing one light intensity image (I(45°,90°)) by applying a partial point spread function corresponding to an area in the raw image obtained by incident of the elliptically polarized light to the image sensor. Image acquisition method. 16. The method of claim 15, wherein obtaining the full Stokes parameter comprises
Constructing full Stokes parameters from four light intensity images (I(0°,0°), I(45°,0°), I(90°,0°), I(45°,90°)) The four Stokes parameters (S ₀ , S ₁ , S ₂ , S ₃ ) are expressed in the equation

Polarized image acquisition method obtained according to.