A single-polarized image based de-scattering method

Abstract

The application discloses a single-polarization-image-based de-scattering method, which is applied to the enhancement of a target image in a turbid water body and comprises the following steps: a time-sharing rotating polarizer is used to acquire a maximum polarization intensity image through a polarization camera; the maximum polarization intensity image is encoded according to a Poincare sphere trajectory, decoded based on a generalized Malus law, and the polarization state of a decoder is adjusted to be orthogonal to an input polarization state to acquire a minimum polarization intensity image; a frequency domain low-pass filter is used to perform frequency domain filtering processing to obtain maximum and minimum backscattering light intensity, and the degree of polarization of backscattering light is calculated; and the clear target object image is calculated by inputting the single-polarization de-scattering model. The method breaks through the limitation of traditional multi-image processing, can realize scattering light suppression only by using a single polarization image, significantly simplifies the complexity of an imaging system, and effectively improves the definition of an underwater image.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of target image enhancement in scattering medium environment, and particularly relates to a single polarization image based de-scattering method. BACKGROUND

[0002] Underwater optical imaging has long been plagued by image quality degradation due to water scattering effects. The absorption and scattering of light by suspended particles result in low contrast, color distortion, and blurred details in underwater images, severely restricting the application of marine exploration, ecological monitoring, and other fields.

[0003] Traditional physical model based de-scattering methods have achieved certain results, but still have the following bottlenecks in actual scenarios: 1) Multi-image dependence and hardware limitations: Methods based on polarization characteristics usually require the acquisition of multiple orthogonal polarization images or Stokes vector images, which have high requirements for the synchronization and stability of imaging devices. 2) Scattering light calculation error: The intensity and polarization of backscattered light often have non-uniform distribution, and traditional polarization de-scattering techniques assume that the relevant information is constant, which can easily lead to loss of target details in complex underwater scenes. 3) Limited use of scenes: Traditional polarization de-scattering techniques rely on the polarization information of backscattered light calculated from the background area, which is difficult to apply to target-dominated scenes.

[0004] In recent years, methods based on polarization physical models have shown significant advantages by separating the polarization characteristics of target light and scattered light. However, existing techniques are still limited by the dependence on multi-angle polarization data and the insufficient accuracy of low-frequency component estimation of backscattered light. SUMMARY

[0005] The problem to be solved by the present application is to provide a single polarization image based de-scattering method, which directly generates orthogonal polarization state images using the generalized Malus law, and realizes frequency domain separation by combining the low-frequency characteristics of backscattered light, thereby reducing hardware complexity, breaking through the limitations of traditional methods in terms of limited scene adaptability and inaccurate scattering light estimation, and improving the imaging quality of the target.

[0006] The present application adopts the following technical solution: A single polarization image based de-scattering method generates orthogonal polarization image pairs using the generalized Malus law, and combines a frequency domain analysis model to suppress backscattering interference, with the following specific steps:

[0007] Step 1, construct a polarization de-scattering imaging system, time-division rotating linear polarizer, and obtain the brightest polarization image as the maximum polarization intensity image through a polarization camera ;

[0008] Step 2, input the maximum polarization intensity image , the decoding result is , the polarization state of the decoder is adjusted according to the coding and decoding principle, and the minimum polarization intensity image is obtained when the polarization state is orthogonal ;

[0009] Step 3, based on the low-frequency property of backscattering light, a frequency domain low-pass filter is used to perform frequency domain filtering processing on the maximum and minimum polarization intensity images and , to obtain the maximum and minimum backscattering light intensity and , and calculate the degree of polarization of backscattering light ;

[0010] Step 4, the degree of polarization of backscattering light is brought into the single-polarization despeckling model to calculate the clear target object image .

[0011] Preferably, the polarization despeckling imaging system in step 1 comprises a light source, a linear polarizer, and a polarization camera.

[0012] The light source emits active illumination light through the linear polarizer to form linearly polarized light, which is irradiated to the target object; the target object is submerged in turbid water containing scattering medium, and the polarization image of the target object under the turbid water is obtained by the polarization camera;

[0013] The linear polarizer is rotated in time sharing, and when the brightness of the polarization image obtained by the polarization camera is maximized, the maximum polarization intensity image is obtained .

[0014] Preferably, in step 2, the input polarization image is encoded according to the Poincare sphere trajectory, the light intensity distribution satisfies the generalized Malus law, and the decoding result is , which is represented as follows:

[0015] ;

[0016] Wherein, represents the total amplitude of incident light, represents the ellipticity of incident light, represents the azimuth angle of incident light, , represents a global parameter.

[0017] Preferably, in step 2, the total amplitude of incident light corresponds to the irradiance received by the polarization camera, the ellipticity of incident light corresponds to the polarization state of the light wave received by the polarization camera, and the azimuth angle of incident light The polarization azimuth of the light wave received by the corresponding polarization camera.

[0018] Preferably, in step 2, the polarization state of the decoder is adjusted to the input polarization state, and the change is The minimum polarization intensity image is obtained, and is represented as follows:

[0019] .

[0020] Preferably, in step 3, the Fourier transform and inverse Fourier transform are used to perform frequency domain filtering on the maximum and minimum polarization intensity images and respectively, to obtain the maximum and minimum backscattering light intensity and , which are represented as follows:

[0021] ;

[0022] ;

[0023] wherein , are the Fourier transform and inverse Fourier transform respectively, is the transfer function of the frequency domain low-pass filter.

[0024] Preferably, in step 3, the degree of polarization of the backscattering light is calculated, and the formula is as follows:

[0025] .

[0026] Preferably, in step 4, a clear target object image is obtained, and is represented as follows:

[0027] .

[0028] The technical scheme of the present application also provides an electronic device comprising:

[0029] one or more processors;

[0030] a storage device having one or more programs stored thereon;

[0031] When the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned any one of the single polarization image-based de-scattering methods.

[0032] The technical solution of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the program implements the steps of any of the above-mentioned descattering methods based on a single polarization image.

[0033] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0034] 1. The method of the present invention innovatively combines the generalized Malus law with frequency domain analysis. Only a single polarization image is required as input to achieve clear imaging in scattering media. There is no need to collect multiple sets of orthogonal polarization images or complex input data such as Stokes vectors, which greatly simplifies the data acquisition process and reduces hardware complexity.

[0035] 2. The method of the present invention directly generates orthogonal polarization state images through the generalized Malus law, and realizes frequency domain separation by combining the low-frequency characteristics of backscattered light. The descattering effect is obvious for target-dominated scenes.

[0036] 3. The method of the present invention is a fully automatic image processing method that does not require background priors and human-computer interaction. It can automatically obtain descattered imaging results, has strong robustness, and has a significant image information enhancement effect. It provides an innovative solution for real-time high-definition underwater imaging and can be applied to marine resource exploration, underwater ecological environment monitoring, lost object salvage and other fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a structural diagram of a polarization descattering imaging system according to an embodiment of the present invention;

[0038] Figure 2 This is a flowchart of the descattering method based on a single polarization image of the present invention;

[0039] Figure 3 Schematic diagram of turbid water underwater intensity imaging results and polarization descattering imaging results based on the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the application are further elaborated in detail below with reference to the accompanying drawings. The described embodiments are only a part of the embodiments involved in the present invention. All non-innovative embodiments of other researchers in this field on this embodiment fall within the scope of protection of the present invention. At the same time, the step numbers in the embodiments of the present invention are only set for the convenience of explanation and description, and the order between the steps is not limited in any way. The execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0041] In one embodiment of the present invention, the polarization descattering imaging system involved in the method of the present invention has a structure as follows: Figure 1As shown, it comprises: a polarization camera 1, an LED light source 2, a linear polarizer 3, a water tank 4, and a target object 5.

[0042] Specifically, the target object 5 is submerged in the water tank 4 filled with scattering medium (whole milk), the light emitted by the LED light source 2 forms linearly polarized light through the linear polarizer 3, and is irradiated on the target object 5 through the water tank 4, and the polarization camera 1 is used to acquire the target object image under turbid water.

[0043] The present application is based on a single polarization image de-scattering method, such as Figure 2 As shown, the specific steps are as follows:

[0044] 1, fix the target object 5 at the bottom of the water tank 4, fix the LED light source 2 and the linear polarizer 3, and rotate the linear polarizer 3 direction in time, so that the polarization image brightness acquired by the polarization camera 1 presents maximization and is saved, and the maximum polarization intensity image is obtained .

[0045] 2, input the maximum polarization intensity image , encode according to the Poincare sphere trajectory, and the polarization light intensity distribution satisfies the generalized Malus law, decode the polarization image encoded by the known Poincare sphere trajectory, and the decoding result is , adjust the image when the polarization state of the decoder is orthogonal to the input polarization state, which is the minimum polarization intensity image .

[0046] 3, using the low-frequency property of backscattering light, the frequency domain analysis method of frequency domain low-pass filter is used to estimate the maximum and minimum backscattering light intensity.

[0047] Specifically, Fourier transform and inverse Fourier transform are used to perform frequency domain filtering processing on the cross-polarization images and , and then the backscattering light and are obtained, and then the degree of polarization of the backscattering light is calculated according to the definition of the degree of polarization.

[0048] 4, input all parameters into the single polarization de-scattering model, calculate the clear target image , and obtain the de-scattering result.

[0049] In this embodiment, experiments in different concentrations of scattering medium are carried out to verify the effectiveness of the method, and the turbidity is taken as an example, as shown in Figure 3 .

[0050] Specifically, the original intensity image of the turbid water acquired by the polarization camera 1 is as shown in Figure 3As shown in (a) in the figure, it can be seen that the target contrast in the image is low, the details are blurred, and the effective information recognition degree is low due to the interference of backscattering light. The result obtained by the polarization despeckling imaging method based on the application is shown in (b) in the figure. Figure 3 As shown in (b) in the figure, it can be seen that the target information degradation effect caused by the scattering light is effectively inhibited by the method of the application, the enhancement effect of the target is realized, and the detail information of the target can be easily identified.

[0051] Particularly, in the single-polarization despeckling model of the embodiment, the underwater optical image is composed of target light, forward scattering light and backscattering light, the forward scattering part causing image blur in the total signal is ignored, and the total intensity Can be described as:

[0052] ;

[0053] Among them, represents the target light, represents the backscattering light.

[0054] The brightest maximum polarization intensity image And the darkest minimum polarization intensity image Respectively are:

[0055] ;

[0056] ;

[0057] Among them, , respectively represent the maximum and minimum target light, , respectively represent the maximum and minimum backscattering light.

[0058] The intensity of the target light and the backscattering light is:

[0059] ;

[0060] ;

[0061] According to the definition of the degree of polarization (DOP), the target light DOP and the backscattering light DOP are:

[0062] ;

[0063] ;

[0064] Among them, , respectively are the target light DOP and the backscattering light DOP.

[0065] Since most of the target light is contained in the minimum target light intensity component is ignored:

[0066] ;

[0067] Therefore, the total intensity is:

[0068] .

[0069] Thus, the clear target image is:

[0070] ;

[0071] Since in this embodiment, can be directly obtained, and according to the low frequency property of backscattering light, the intensity and degree of polarization of backscattering light can be estimated according to the maximum and minimum polarization intensity images using frequency domain analysis method. Therefore, the key is to obtain the minimum polarization intensity image, i.e. the cross linear polarization image.

[0072] Further, according to the generalized Malus law, the decoding result of the polarization image which has encoded the known Poincare sphere trajectory is:

[0073] ;

[0074] wherein, is the ellipticity of the incident light, is the azimuth angle of the incident light, which are used to describe the elliptical shape and spatial orientation of the polarization state respectively; is the total amplitude of the incident light.

[0075] In the imaging process, corresponds to the irradiance received by the camera, corresponds to the polarization state of the light wave received by the camera, corresponds to the polarization azimuth angle of the light wave received by the camera.

[0076] and are global parameters, and the calculation formulas are respectively:

[0077] ;

[0078] ;

[0079] wherein, and are given encoding trajectories:

[0080] ;

[0081] .

[0082] Therefore, changing for , can generate , expressed as:

[0083] .

[0084] Further, to achieve the enhancement of the target image and avoid human-computer interaction, the estimation and removal of backscattering light are needed, since the backscattering light has low frequency properties, the embodiment adopts a frequency domain low-pass filter to estimate the intensity of the backscattering light:

[0085] ;

[0086] ;

[0087] Wherein, and respectively represent Fourier transform and inverse Fourier transform, is the transfer function of the frequency domain low-pass filter.

[0088] Preferably, is expressed as follows:

[0089] ;

[0090] Wherein, represents the distance from the frequency origin, is the standard deviation of the Gaussian distribution.

[0091] Further, the backscattering light DOP is updated as:

[0092] ;

[0093] The clear target image is updated as:

[0094] .

[0095] It can be seen that the present application proposes a kind of depolarization imaging method based on single polarized image to the restriction that traditional polarization imaging needs multiple polarization azimuth angle image input, combine generalized Malus law with frequency domain analysis, without background and prior knowledge, only single polarized input can realize clear imaging in scattering medium, and remove backscattering light by frequency domain filtering, in turbid water and other scattering medium environment, the contrast and recognition degree of target image are significantly improved, and finally the enhancement of target image in scattering medium is realized.The present application method is simple and universal, robust, and the image information recovery effect is remarkable, can be widely applied to underwater detection, ocean engineering monitoring and underwater robot vision system.

[0096] The embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon; and when the one or more programs are executed by the one or more processors, the one or more processors implement the single-polarization image-based de-scattering imaging method described in the above embodiment.

[0097] The embodiment of the present application also provides a computer readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the single-polarization image-based de-scattering imaging method in the above embodiment are implemented.

[0098] The above merely describes the preferred embodiments of the present application, and it should be noted that those of ordinary skill in the art can make several improvements and refinements without departing from the principles of the present application, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A descattering method based on a single polarization image, characterized in that: The generalized Malus law is used to generate orthogonal polarization image pairs, and the frequency domain analysis model is combined to suppress backscatter interference. The specific steps are as follows: Step 1: Build a polarization descattering imaging system, rotate the linear polarizer in time, and obtain the brightest polarization image through the polarization camera as the maximum polarization intensity image. ; Step 2: Input the maximum polarization intensity image , encoded according to the Poincare sphere trajectory and decoded based on the generalized Malus law, the decoding result is , according to the coding principle, adjust the polarization state of the decoder and the input polarization state, and obtain the minimum polarization intensity image when they are orthogonal ; Input polarization image , encoded according to the Poincare sphere trajectory, the light intensity distribution satisfies the generalized Malus law, and the decoding result is , which is expressed as follows: ； in, represents the total amplitude of the incident light, represents the ellipticity of the incident light, represents the azimuth of the incident light, 、 Represents global parameters; Adjust the polarization state of the decoder and the input polarization state to change for When the minimum polarization intensity image is obtained , which is expressed as follows: ； Step 3: Based on the low-frequency nature of the backscattered light, a frequency domain low-pass filter is used to filter the maximum and minimum polarization intensity images. and Perform frequency domain filtering to obtain the maximum and minimum backscattered light intensities and , and calculate the polarization degree of the backscattered light ; Step 4: Polarization of backscattered light Bring in the single polarization descattering model to calculate a clear image of the target object .

2. The descattering method based on a single polarization image according to claim 1, characterized in that: The polarization descattering imaging system includes: a light source, a linear polarizer, and a polarization camera; The light source emits illumination light that passes through a linear polarizer to form linearly polarized light, which is then irradiated onto a target object; the target object is submerged in turbid water containing a scattering medium, and a polarization image of the target object under the turbid water is obtained by the polarization camera; The linear polarizer is rotated in time, a polarization image is acquired by a polarization camera, and the brightest polarization image is selected as the maximum polarization intensity image. .

3. The descattering method based on a single polarization image according to claim 1, characterized in that: In step 2, the total amplitude of the incident light The ellipticity of the incident light corresponds to the irradiance received by the polarization camera. The polarization state of the light wave received by the polarization camera corresponds to the azimuth angle of the incident light Corresponding to the polarization azimuth of the light wave received by the polarization camera.

4. The descattering method based on a single polarization image according to claim 1, characterized in that: In step 3, the maximum and minimum polarization intensity images are transformed using Fourier transform and inverse Fourier transform respectively. and Perform frequency domain filtering to obtain the maximum and minimum backscattered light intensities and , which is expressed as follows: ； ； in, 、 are Fourier transform and inverse Fourier transform, respectively. is the transfer function of the frequency domain low-pass filter.

5. The descattering method based on a single polarization image according to claim 4, characterized in that: In step 3, calculate the polarization degree of the backscattered light , the formula is as follows: 。 6. The descattering method based on a single polarization image according to claim 5, characterized in that: In step 4, a clear image of the target object is obtained based on the single polarization descattering model. , expressed as: 。 7. An electronic device, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the descattering method based on a single polarization image as claimed in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that A computer program is stored thereon, and when the program is executed by a processor, the steps of the descattering method based on a single polarization image according to any one of claims 1 to 6 are implemented.

Images