A water surface glare automatic suppression method based on a focal plane polarization camera

Abstract

The application discloses a water surface glare automatic suppression method based on a split focal plane polarized camera and belongs to the field of water surface environment detection. The application obtains four polarized radiation images with different polarization directions of 0 DEG, 45 DEG, 90 DEG and 135 DEG of a scene at the same time through the split focal plane polarized camera, independently calculates the polarization information such as Stokes vector, polarization degree and deflection angle of different regions on the image, acquires the polarization information of dynamic scenes and moving targets at the same time, and the processing result does not cause problems such as trailing, misplacement and blur; based on the polarization imaging principle, the best polarization angle corresponding to the minimum average gray scale and the polarization image are calculated, the suppression of the water surface glare is realized, and based on the non-uniformity of each point of the image, the better suppression effect of the water surface glare is obtained through pixel-by-pixel processing. The application is suitable for the fields of polarization imaging, water body image processing, security monitoring and the like, and improves the definition of target details and contour information in the water body.

Description

TECHNICAL FIELD

[0001] The present application relates to a water surface glare automatic suppression method based on a focal plane polarization camera, belonging to the field of water surface environment detection. BACKGROUND

[0002] Improvement of water environment not only can realize the harmonious development of man and nature, but also can improve people's quality of life and develop regional economy, which has very important practical significance for human survival and development. Water environment monitoring tasks mainly include water surface floating objects, aquatic organisms, underwater plants and underwater pollutants. China has extensive rivers, lakes, especially the coastal areas of several major rivers, delta tributaries and lakes, and the industrial economy of the region is developed, so the environmental monitoring task is arduous. With the progress of technology, not only various cameras on traditional water surface monitoring ships, but also fixed-point monitoring cameras along the river or unmanned dynamic cameras have become important monitoring means, which can quickly and effectively distinguish the monitored water area combined with various image artificial intelligence processing. However, due to the reflection of the sky and sunlight on the water surface, strong radiation interference, i.e. water surface solar glare, is formed, which is a strong radiation interference in marine water color remote sensing, water surface target monitoring and other water body image detection applications, and is easy to cause problems such as large-area pixel saturation of the imaging detector, loss of pixel information and interference with underwater target detection. Therefore, how to eliminate or reduce the influence of water surface glare reflection on imaging is a key link to improve environmental monitoring.

[0003] Water surface glare has obvious polarization characteristics, and polarization images can reduce the influence of solar glare on rough sea surface. When the wind blows over the water surface, it forms ripples with different degrees of inclination or distortion, causing regional distribution of glare. Therefore, the polarization characteristics of the glare are related to the solar azimuth and the observation conditions of the water surface. By installing a rotating linear polarizer in front of the camera lens, a time-sharing polarization imaging system can be formed to obtain polarization images and polarization information (polarization degree, polarization angle, Stokes vector, etc.). However, in practical applications, the time-sharing polarization imaging system has a complex structure, and the platform and detection points change constantly in practical applications, so the best polarization direction needs to be dynamically optimized and adjusted, and the relative inclination of the water surface in the detection field of view is also inconsistent. Therefore, it is difficult to accurately adjust the points dynamically. The time-sharing polarization imaging system can only take the same polarization direction for the entire image, but for different regions in the image, using only a single polarizer cannot obtain images with the best polarization direction for each part. More importantly, the rotating angle accuracy of the time-sharing system is low, the polarization detection capability is poor, and it is only suitable for static target scenes. For scenes with moving targets, it is difficult to obtain good results, and it is difficult to adapt to the suppression of water surface glare and improve environmental monitoring applications. SUMMARY

[0004] To address the problems of low rotation angle accuracy, poor polarization detection capabilities, and applicability to static target scenes in time-sharing polarization imaging systems, this invention provides an automatic water surface glare suppression method based on a split-focus plane polarization camera. Based on the principles of polarization imaging, the method calculates the optimal polarization angle and polarization image corresponding to the minimum average grayscale, suppressing water surface glare and improving the clarity of target details and contours in the water.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] The present invention discloses a method for automatically suppressing water surface glare based on a split-focus plane polarization camera. Based on the significant partial polarization characteristics of water surface glare, the split-focus plane polarization camera is used to directly obtain images of a scene in different polarization directions at the same time. Polarization measurements are then performed on the water surface affected by glare, obtaining polarization radiation patterns at four polarization directions: 0°, 45°, 90°, and 135°, and calculating their Stokes parameters. Based on the principle of polarization imaging, the optimal polarization angle and polarization image corresponding to the minimum average grayscale are calculated to suppress water surface glare. Furthermore, based on the inhomogeneity of each point in the image, pixel-by-pixel processing is performed to achieve a better suppression effect on water surface glare, making target details and contour information in the water clearer.

[0007] The present invention discloses a method for automatically suppressing water surface glare based on a split-focus plane polarization camera, comprising the following steps:

[0008] Step 1: Use a split-focus plane polarization camera to capture polarization images of the water surface affected by glare;

[0009] The polarization image is obtained by interpolation method using polarization information of four different polarization angles of 0°, 45°, 90° and 135° at the same time and in the same scene acquired by a split-focus plane polarization camera.

[0010] Step 2: Calculate polarization information;

[0011] According to the principle of polarization imaging, the polarization information is calculated using the intensity of the polarization images in the four directions obtained in step 1, including the line Stokes vector, polarization degree, and polarization angle of the image;

[0012] Step 3: Calculate the light intensity value at each polarization angle;

[0013] According to the polarization imaging model, the Stokes vector obtained in step 2 is used to calculate the light intensity value of the scene at each polarization angle in theory.

[0014] Step 4: Calculate the image in the optimal polarization direction;

[0015] The gray scale average value is used as an index for evaluating the influence of the intensity of the interference light on the image, the image intensity values of the scene calculated in step three in each polarization direction are compared, if the image gray scale average value in the direction of θ i is the minimum, then θ i is the optimal angle θ best for suppressing the sun glint. best i

[0016]

[0017]

[0018] The image in this direction is represented as:

[0019]

[0020] Wherein, S0, S1 and S2 are Stokes vectors, at this time, the image gray scale average value is the minimum, and the high-intensity glint in the image is filtered to the maximum extent.

[0021] Considering that the polarization characteristics of each part of the dynamic water surface are not completely the same, the image is processed pixel by pixel, the Stokes vector at (i, j) is denoted as S(i, j), and the corresponding Mueller matrix is denoted as M m(i,j) , and the pixel value after polarization processing is

[0022] I θ(i,j) = M m(i,j) · S(i, j)

[0023] Step five: the image after the above polarization processing is subjected to histogram equalization to obtain the final result, so that the suppression of the water surface glint is realized, and the clarity of the target details and contour information in the water body is improved.

[0024] Advantages:

[0025] 1. The water surface glint automatic suppression method based on a split focal plane polarization camera disclosed in the present application uses a split focal plane polarization camera and calculates the polarization image in the optimal polarization direction to suppress the water surface glint for water surface scene detection according to the characteristics that the reflected glint part of the water surface scene has strong polarization intensity, and the non-glint part has weak polarization, without manual or mechanical rotation of the polarization plate, so that mechanical errors caused in the process are reduced, and the clarity of the target details and contour information in the water body is improved.

[0026] ​​​2. This invention discloses an automatic water surface glare suppression method based on a split-focus plane polarization camera. This method uses a split-focus plane polarization camera to capture polarization images of a water scene simultaneously in four different polarization directions. The polarization image in the optimal polarization direction for suppressing water surface glare is then calculated. Compared to time-sharing systems, polarization information for dynamic scenes and moving targets is acquired simultaneously, resulting in a processing method that avoids the smearing, misalignment, and blurring that plague time-sharing systems. Furthermore, this method is suitable for long-term continuous operations and has great application potential.

[0027] 3. Compared with the time-sharing polarization imaging system, the present invention discloses a method for automatically suppressing water surface glare based on a split-focus plane polarization camera. The split-focus plane polarization camera simultaneously acquires four images of the scene at the same moment, with different polarization directions at 0°, 45°, 90°, and 135°. The method can independently calculate the polarization information such as the Stokes vector, degree of polarization, and deflection angle of different regions on the image. The polarization direction with the minimum average grayscale and the intensity image in this direction are calculated, and the real-time changing water surface is processed pixel by pixel. This method is more suitable for actual water surface scenes with fluctuations due to wind or object movement, and is more effective in suppressing water surface glare, thereby improving the clarity of target details and contour information in the water body. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 The present invention discloses a flow chart of a method for automatically suppressing water surface glare based on a split-focus plane polarization camera;

[0029] Figure 2 This is a diagram showing the application results of a method for automatically suppressing water surface glare based on a split-focus plane polarization camera disclosed in this embodiment.

[0030] Figure (a) is the original water surface image, Figure (b) is the image of the original image processed by the method of the present invention; Figure (c) is the image after equalization processing is performed on Figure (d). DETAILED DESCRIPTION

[0031] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. The technical problems solved by the technical solution of the present invention and the beneficial effects thereof are also described. It should be noted that the described embodiments are only intended to facilitate understanding of the present invention and do not serve to limit the present invention in any way.

[0032] like Figure 1 As shown, this embodiment discloses a method for automatically suppressing water surface glare based on a focal plane polarization camera, which is specifically applied to water surface monitoring in Shichahai, including the following steps:

[0033] Step one: Collecting image data, using Phoenix PHX050S-P split focal plane visible light polarization camera to collect the polarization images of water surface affected by glare, obtaining the polarization information of the same scene at the same time under different polarization angles (0°, 45°, 90° and 135°), and calculating the polarization images in each direction by interpolation method. As shown in Fig. Figure 2 (a) is the intensity image of three water surface scenes.

[0034] Step two: Calculate the polarization information. According to the principle of polarization imaging, calculate the image line Stokes vector I0, I 45 , I 90 , I 135 , degree of polarization DoP, polarization angle AoP and other polarization information:

[0035] S0=I0+I 90

[0036] S1=I0-I 90

[0037] S2=I 45 -I 135

[0038]

[0039] Step three: According to the polarization imaging model, the change of the polarization state of light wave by optical element is described by Mueller matrix M:

[0040]

[0041] Where, S in is the Stokes vector of incident light; S out is the Stokes vector of outgoing light; M is the Mueller matrix of optical element, which represents the effect of optical element or system on incident light.

[0042] Since photoelectric imaging device can only respond to light intensity, the total light intensity I detected by photoelectric imaging device is:

[0043] I=M 11 ·S 0_in +M 12 ·S 1_in +M 13 ·S 2_in +M 14 ·S 3_in

[0044] The general expression of Mueller matrix M p of ideal linear polarizer with angle θ between transmission and horizontal direction (x axis) is:

[0045]

[0046] Remember M p The first line of M m , the analyzed image after passing through the polarizer with an analyzer angle θ can be expressed as:

[0047]

[0048] Polarized light intensity I θ It is related to the transmission direction θ of the linearly polarized light. When the transmission direction θ is perpendicular to the vibration direction α of the polarized light, the intensity of the analyzed polarized light reaches the minimum value and the glare is suppressed to the greatest extent.

[0049] Step 4: Use the grayscale average as an indicator to evaluate the effect of the image on the intensity of the interfering light:

[0050] For the image f with an analyzer angle θ θ (i,j), defines the grayscale average value As an evaluation indicator of interference from water surface glare:

[0051]

[0052] Where WH is the size of the evaluation image area.

[0053] If the detector angle θ i The average grayscale value of the image in the direction is the minimum, then θ i The optimal angle θ for sun glare suppression best :

[0054]

[0055] θ best =θ i

[0056] The obtained optimal angle image can be expressed as:

[0057]

[0058] At this time, the average grayscale value of the image is the smallest, and the high-intensity glare in the image is filtered out to the greatest extent.

[0059] In reality, the water surface is not absolutely still due to the influence of moving objects and ambient wind. Considering that the polarization characteristics of each part are not exactly the same, in order to obtain better image effects, the polarization solution is performed for each pixel. Let the Stokes vector at (i, j) be S(i, j), and the corresponding Mueller matrix be M m(i,j) , then the light intensity after polarization processing is

[0060] I θ(i,j) =M m(i,j)• S(i,j)

[0061] As Figure 2 (b) shows the results of polarization processing under three water surface scenes.

[0062] Step five: histogram equalization is performed on the polarization-processed images above to obtain the final results. As Figure 2 (c) shows the results of histogram equalization after polarization processing under three water surface scenes. The suppression of water surface glare and the improvement of the clarity of target details and contour information in the water body are achieved.

[0063] The above detailed description further illustrates the purpose, technical solutions and beneficial effects of the application. It should be understood that the above description is only a specific embodiment of the application and is not intended to limit the protection scope of the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the application shall be included in the protection scope of the application.

Claims

1. A method for automatic suppression of water surface glint based on a focal plane polarized camera, characterized in that: The method comprises the following steps: Step one: acquire the polarized image of water surface affected by glare by using a focal plane polarized camera; The polarized image is the polarized information of the same scene at the same time under four different polarized angles of 0°, 45°, 90° and 135° acquired by the focal plane polarized camera, and is obtained by interpolation method; Step two: calculate the polarized information; According to the principle of polarized imaging, the intensity of the four direction polarized images obtained in step one is used to calculate the polarized information, including the linear Stokes vector, the degree of polarization and the polarization angle of the image; Step three: calculate the light intensity value on each polarization angle; According to the polarized imaging model, the Stokes vector obtained in step two is used to calculate the light intensity value of the scene on each polarization angle in theory; Step four: calculate the image on the best polarization direction; The grayscale average is used as an indicator to evaluate the influence of the intensity of the interfering light on the image. The image intensity values ​​of the scene in each polarization direction calculated in step 3 are compared. If the analyzer angle θ i The average grayscale value of the image in the direction is the minimum, then θ i The optimal angle θ for sun glare suppression best : θ best = θ i The image on this direction is represented as: Wherein, S0, S1 and S2 are Stokes vectors; at this time, the average value of the image gray scale is minimum, and the high-intensity glare in the image is filtered to the maximum extent; Considering that the polarization characteristics of different parts of the dynamic water surface are not completely the same, the image is processed pixel by pixel; let the Stokes vector at (i, j) be S(i, j), and the corresponding Mueller matrix be M m(i,j) The pixel value after polarization processing is I θ(i,j) = M m(i,j) · S(i,j) Step five: perform histogram equalization on the polarized image to obtain the final result, so as to realize the suppression of the water surface glare and improve the clarity of the target details and contour information in the water body.

Images