Electromagnetic wave polarization imaging fusion enhancement method and system

Abstract

The electromagnetic wave polarization imaging fusion enhancement method and system belong to the field of remote sensing and detection technology. The problem that the target and background contrast effect is poor and the image quality is affected in the existing fusion process of different polarization images is solved. The present invention obtains four polarization images of the target detection scene; the four polarization images are: horizontal polarization image, 45-degree linear polarization image, vertical polarization image and 135-degree linear polarization image; according to the polarization angle relationship of the four polarization images, the four polarization images are converted into Stokes' second component TQ image and third component TU image by using the pixel values ​​of the pixel points in the four polarization images; the Stokes' second component TQ image and third component TU image are converted into four regional images; and the mark of each regional image is obtained; using the mark of each regional image, the four regional images are fused to obtain an enhanced image. The present invention is suitable for the fusion enhancement of electromagnetic wave multi-polarization imaging of target detection scenes.

Description

Technical Field

[0001] The present invention belongs to the field of remote sensing and detection technology, and in particular relates to a method and system for fusing electromagnetic wave polarization imaging to obtain an enhanced image. Background Art

[0002] Microwave / millimeter wave / terahertz radiation imaging is a typical electromagnetic wave radiation imaging. Its system realizes remote sensing and detection of the observation scene by receiving microwave / millimeter wave / terahertz thermal radiation spontaneously or reflected by materials in nature. It has the advantages of working around the clock, quasi-all-weather, good concealment, no radiation hazard and the ability to penetrate a variety of coverings. Therefore, it has been applied to the fields of atmospheric remote sensing, ocean monitoring, soil and vegetation remote sensing, human security inspection, sea surface target detection, lunar and planetary remote sensing, key site monitoring, military detection, etc.

[0003] In the field of target detection such as human security inspection and military detection, the grayscale contrast in electromagnetic wave radiation images is usually used to detect and identify targets. Existing technologies are still difficult to popularize on a large scale, mainly because the system hardware performance parameters (such as spatial resolution, temperature sensitivity, etc.) and the performance of detection and recognition algorithms need to be improved. Under given hardware performance conditions, how to improve the contrast between the target and the background is of great significance. Traditional imaging methods generally use a single polarization to image the observation scene, and the information obtained is limited. Multi-polarization imaging can obtain more information and is expected to improve detection and recognition capabilities. Studies have found that in images of different polarizations, there is a significant difference in the contrast between local targets and the background, which not only brings difficulties to target detection, but also affects the visual perception of the image and interferes with the observer. Any single polarization image cannot obtain the optimal contrast performance of all local targets at the same time, that is, no polarization has an absolute advantage over other polarizations. Therefore, how to intelligently fuse local areas with optimal contrast performance in multiple polarization images is one of the key issues that need to be solved in order to use multi-polarization imaging to improve target detection capabilities. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that the contrast between target and background is poor and the image quality is affected in the existing fusion process of different polarization images, and an electromagnetic wave polarization imaging fusion enhancement method and system are proposed.

[0005] The electromagnetic wave polarization imaging fusion enhancement method of the present invention comprises:

[0006] Step 1: Acquire four polarization images of the target detection scene; the four polarization images are: a horizontal polarization image, a 45-degree linear polarization image, a vertical polarization image, and a 135-degree linear polarization image;

[0007] Step 2: According to the polarization angle relationship of the four polarization images, the four polarization images are converted into the second component TQ image and the third component TU image of Stokes by using the pixel values ​​of the pixel points in the four polarization images;

[0008] Step 3: convert the second component TQ image and the third component TU image of Stokes into four regional images; and obtain the mark of each regional image;

[0009] Step 4: Using the markers of each region image, fuse the four region images to obtain an enhanced image.

[0010] Furthermore, in the present invention, step one also includes a process of denoising the four polarization images.

[0011] Furthermore, in the present invention, in step 2, the method for obtaining the pixel values ​​of the pixel points in the four polarization images is: converting the four polarization images into two-dimensional matrix forms respectively, and obtaining the pixel value of each pixel point in the polarization image.

[0012] Furthermore, in the present invention, in step 2, the specific method of converting the four polarization images into the second component TQ image and the third component TU image of Stokes is:

[0013] Using the formula:

[0014] TQ(i)=T1(i)-T3(i),

[0015] TU(i)=T2(i)-T4(i),

[0016] The Stokes second component TQ image and the third component TU image are calculated and obtained, where i is the pixel index, TQ(i) is the pixel value of the i-th pixel of the second component TQ image, TU(i) is the pixel value of the i-th pixel of the third component TU image, T1(i) is the pixel value of the i-th pixel of the horizontal polarization image, T2(i) is the pixel value of the i-th pixel of the 45-degree linear polarization image, T3(i) is the pixel value of the i-th pixel of the vertical polarization image, and T4(i) is the pixel value of the i-th pixel of the 135-degree linear polarization image.

[0017] Furthermore, in the present invention, in step 3, the specific method of converting the second component TQ image and the third component TU image of Stokes into four regional images is:

[0018] pass:

[0019]

[0020]

[0021]

[0022]

[0023] Four regional images are obtained, where TS1(i) is the i-th pixel image of the first sub-region, TS2(i) is the i-th pixel image of the second sub-region, TS3(i) is the i-th pixel image of the third sub-region, and TS4(i) is the i-th pixel image of the fourth sub-region; |*| is an absolute value operator, and the pixel index with a pixel value equal to 1 in the four regional images is the mark position information.

[0024] Furthermore, in the present invention, in step 4, the specific method of fusing the four regional images to obtain the enhanced image using the label of each regional image is as follows:

[0025] By formula:

[0026]

[0027] The four-region image fusion is realized, where TF(i) is the pixel value of the i-th pixel of the fused enhanced image TF.

[0028] An electromagnetic wave polarization imaging fusion enhancement system, used to execute the above-mentioned electromagnetic wave polarization imaging fusion enhancement method, the system comprises: a focusing lens, a polarization antenna, a radiometer channel and a data acquisition and processing device;

[0029] The focusing lens is used to focus the electromagnetic waves spontaneously radiated or reflected from the target detection scene onto the polarized antenna;

[0030] The polarized antenna and the radiometer channel are connected to form an image observation unit to collect horizontal polarization images, 45-degree linear polarization images, vertical polarization images and 135-degree linear polarization images of the target detection scene;

[0031] The data acquisition and processing device converts the four polarization images into the second component TQ image and the third component TU image of Stokes; then converts the second component TQ image and the third component TU image into four regional images; and finally fuses the four regional images to obtain an enhanced image.

[0032] Furthermore, in the present invention, the data acquisition and processing device includes a data collector, a data processor and a display;

[0033] The data collector collects four polarization images sent by the radiometer channel and sends the four polarization images to the data processor;

[0034] The data processor is used to convert the four polarization images into Stokes second component TQ images and third component TU images after denoising, and then convert the second component TQ images and third component TU images into four regional images; finally, the four regional images are fused to obtain an enhanced image;

[0035] The display is used to display an enhanced image.

[0036] Furthermore, in the present invention, there is one image observation unit, which is arranged on an observation axis, and the image observation unit rotates around the observation axis to collect four polarization images in a time-sharing manner.

[0037] Furthermore, in the present invention, there are four image observation units, and the four image observation units collect four polarization images simultaneously.

[0038] The method of the present invention can fuse the optimal local areas in different polarization images, so that all targets and backgrounds in the fused single image have optimal contrast, enhance image quality, and thus improve target detection capability and image visual perception. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a flow chart of the method of the present invention;

[0040] Figure 2 It is the principle block diagram of the electromagnetic wave polarization imaging system;

[0041] Figure 3 are four polarization images in a specific embodiment;

[0042] Figure 4 are the Stokes second component TQ image and the third component TU image in a specific embodiment;

[0043] Figure 5 are four region images in a specific embodiment;

[0044] Figure 6 is an enhanced image in a specific embodiment. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0047] Specific implementation method 1: The following is combined Figure 1 This embodiment describes the remote sensing imaging polarization fusion enhancement method described in this embodiment, including:

[0048] Step 1: Acquire four polarization images of the target detection scene; the four polarization images are: a horizontal polarization image, a 45-degree linear polarization image, a vertical polarization image, and a 135-degree linear polarization image;

[0049] Step 2: According to the polarization angle relationship of the four polarization images, the four polarization images are converted into the second component TQ image and the third component TU image of Stokes by using the pixel values ​​of the pixel points in the four polarization images;

[0050] Step 3: convert the second component TQ image and the third component TU image of Stokes into four regional images; and obtain the mark of each regional image;

[0051] Step 4: Using the markers of each region image, fuse the four region images to obtain an enhanced image.

[0052] The multi-polarization electromagnetic wave imaging system in this embodiment can be implemented by using a multi-polarization microwave, millimeter wave or terahertz imaging system.

[0053] Furthermore, in the present invention, step one also includes a process of denoising the four polarization images.

[0054] Furthermore, in the present invention, in step 2, the method for obtaining the pixel values ​​of the pixel points in the four polarization images is: converting the four polarization images into two-dimensional matrix forms respectively, and obtaining the pixel value of each pixel point in the polarization image.

[0055] Furthermore, in the present invention, in step 4, the specific method of converting the four denoised images into the Stokes second component TQ image and the third component TU image by using the pixel value of each pixel point in the four denoised images is:

[0056] use:

[0057] TQ(i)=T1(i)-T3(i),

[0058] TU(i)=T2(i)-T4(i),

[0059] The Stokes second component TQ image and the third component TU image are calculated and obtained, where i is the pixel index, TQ(i) is the pixel value of the i-th pixel of the second component TQ image, TU(i) is the pixel value of the i-th pixel of the third component TU image, T1(i) is the pixel value of the i-th pixel of the horizontal polarization image, T2(i) is the pixel value of the i-th pixel of the 45-degree linear polarization image, T3(i) is the pixel value of the i-th pixel of the vertical polarization image, and T4(i) is the pixel value of the i-th pixel of the 135-degree linear polarization image.

[0060] Furthermore, in the present invention, in step 3, the specific method of converting the second component TQ image and the third component TU image of Stokes into four regional images is:

[0061] pass:

[0062]

[0063]

[0064]

[0065]

[0066] Four regional images are obtained, where TS1(i) is the i-th pixel image of the first sub-region, TS2(i) is the i-th pixel image of the second sub-region, TS3(i) is the i-th pixel image of the third sub-region, and TS4(i) is the i-th pixel image of the fourth sub-region; |*| is an absolute value operator, and the pixel index with a pixel value equal to 1 in the four regional images is the mark position information.

[0067] Furthermore, in the present invention, in step 4, the specific method of fusing the four regional images to obtain the enhanced image using the label of each regional image is as follows:

[0068] By formula:

[0069]

[0070] The four-region image fusion is realized, where TF(i) is the pixel value of the i-th pixel of the fused enhanced image TF.

[0071] Specific implementation method 2: Combination Figure 2 This embodiment is described. The electromagnetic wave polarization imaging fusion enhancement system described in this embodiment is used to execute the electromagnetic wave polarization imaging fusion enhancement method described in the specific embodiment 1. The system includes: a focusing lens 2, a polarization antenna 3, a radiometer channel 4 and a data acquisition and processing device 5;

[0072] The focusing lens 2 is used to focus the electromagnetic waves spontaneously radiated or reflected by the target detection scene 1 onto the polarized antenna 3;

[0073] The polarized antenna 3 and the radiometer channel 4 are connected to form an image observation unit to collect horizontal polarization images, 45-degree linear polarization images, vertical polarization images and 135-degree linear polarization images of the target detection scene;

[0074] The data acquisition and processing device 5 converts the four polarization images into the second component TQ image and the third component TU image of Stokes; then converts the second component TQ image and the third component TU image into four regional images; and finally fuses the four regional images to obtain an enhanced image.

[0075] Furthermore, in the present invention, the data acquisition and processing device 5 includes a data collector 51, a data processor 52 and a display 53;

[0076] The data collector 51 collects four polarization images sent by the radiometer channel 4, and sends the four polarization images to the data processor 52;

[0077] The data processor 52 is used to convert the four polarization images into Stokes second component TQ images and third component TU images after denoising, and then convert the second component TQ images and third component TU images into four regional images; finally, the four regional images are fused to obtain an enhanced image;

[0078] The display 53 is used to display the enhanced image.

[0079] Furthermore, in the present invention, there is one image observation unit, which is arranged on an observation axis, and the image observation unit rotates around the observation axis to collect four polarization images in a time-sharing manner.

[0080] Furthermore, in the present invention, there are four image observation units, and the four image observation units collect four polarization images simultaneously.

[0081] The method of the present invention fuses the best local areas in different polarization images, so that all target objects and human background in the fused single image have the best contrast, enhances the image quality, and thus improves the target detection capability and image visual perception.

[0082] Combination Figure 3 to Figure 6 Describe this embodiment, specific embodiment:

[0083] Taking the human body security inspection imaging of a person in a typical indoor environment as an example, a millimeter wave imaging system is used to obtain four polarization images, which are finally fused to obtain an enhanced image. In the millimeter wave imaging, the contrast of each object in different linear polarization images is different. Through the above processing method, the optimal local areas in different polarization images in the image are fused, so that all target objects and the human body background in the fused single image have the optimal contrast.

[0084] First, a millimeter wave multi-polarization imaging system is used to obtain horizontal polarization, 45 degree linear polarization, vertical polarization, and 135 degree linear polarization images. In this embodiment, the four polarization images obtained are as follows: Figure 3 As shown, Figure 3 From left to right in the figure are horizontal polarization image T1, 45-degree linear polarization image T2, vertical polarization image T3, and 135-degree linear polarization image T4. It can be seen that the knife on the edge of the chest has the best contrast in the T1 image, but is almost invisible in the T3 image; the object on the left shoulder is very blurry in the T1 image, but has the best contrast in the T3 image. Therefore, intelligently selecting local areas of different polarization images for fusion can make all target objects and human background in the fused single image have the best contrast to enhance the image quality. Before processing, the original image is denoised using filtering algorithms such as Wiener filtering and median filtering.

[0085] Then, the pixel values ​​of the Stokes second component TQ and third component TU images are obtained by the following formula:

[0086] TQ(i)=T1(i)-T3(i),

[0087] TU(i)=T2(i)-T4(i),

[0088] Wherein, i is the pixel index, TQ(i) is the pixel value of the i-th pixel of the second component TQ image, TU(i) is the pixel value of the i-th pixel of the third component TU image, T1(i) is the pixel value of the i-th pixel of the horizontal polarization image, T2(i) is the pixel value of the i-th pixel of the 45-degree linear polarization image, T3(i) is the pixel value of the i-th pixel of the vertical polarization image, and T4(i) is the pixel value of the i-th pixel of the 135-degree linear polarization image.

[0089] The conversion in this step is realized by a computing device. When processing, the horizontal polarization, 45-degree linear polarization, vertical polarization, and 135-degree linear polarization images are respectively substituted into the formula in matrix form for calculation to obtain the TQ and TU images, such as Figure 4 Similarly, the subsequent calculation process is also performed in the form of a matrix image.

[0090] Afterwards, according to the mathematical relationship between the pixel values ​​of the TQ image and the pixel values ​​of the TU image, four sub-region images of the observed scene are obtained. In order to distinguish the pixels of the human body area and the background, a common threshold method such as the Otsu method can be used to remove all pixels other than the human body in all images. The first sub-region image TS1, the second sub-region image TS2, the third sub-region image TS3 and the fourth sub-region image TS4 are obtained respectively by the following four formulas, as shown in FIG. Figure 5 shown.

[0091]

[0092]

[0093]

[0094]

[0095] Among them, |*| is the absolute value operator.

[0096] Finally, according to the marker position information in the four sub-region images, the four polarization images are fused to obtain an enhanced image, such as Figure 6 As shown. The calculation formula for the enhanced image TF obtained by fusing the four polarization images is:

[0097]

[0098] Among them, TF(i) is the pixel value of the i-th pixel of the fused enhanced image TF, and the pixel index with a pixel value equal to 1 in the TS1, TS2, TS3 and TS4 images is the marker position information.

[0099] At this point, all the work of multi-polarization imaging and polarization fusion enhancement has been completed. The final image has completed the fusion of the optimal local areas in the different polarization images of human body security inspection, so that all hidden object targets and human body backgrounds in the fused single image have optimal contrast, enhancing the image quality, thereby improving the target detection capability and image visual perception.

[0100] Contrast enhancement between targets and backgrounds in images is of great significance for electromagnetic wave imaging applications. The electromagnetic wave polarization imaging fusion enhancement method and system provided by the present invention obtains more information about the scene by using the multi-polarization imaging system to obtain multi-polarization images, makes full use of multi-polarization information, breaks through the limitation that some targets in traditional single polarization images cannot obtain optimal contrast, fuses the optimal local areas in different polarization images, and makes all targets and backgrounds in the fused single image have optimal contrast. The processing method is simple and effective, and has good robustness.

[0101] Although the present invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. It should therefore be understood that many modifications may be made to the exemplary embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the various dependent claims and features described herein may be combined in a manner different from that described in the original claims. It should also be understood that features described in conjunction with individual embodiments may be used in other described embodiments.

Claims

1. A method for fusion enhancement of electromagnetic wave polarization imaging, characterized in that: include: Step 1: Acquire four polarization images of the target detection scene; the four polarization images are: a horizontal polarization image, a 45-degree linear polarization image, a vertical polarization image, and a 135-degree linear polarization image; Step 2: According to the polarization angle relationship of the four polarization images, the four polarization images are converted into the second component TQ image and the third component TU image of Stokes by using the pixel values ​​of the pixel points in the four polarization images; Step 3: convert the second component TQ image and the third component TU image of Stokes into four regional images; and obtain the mark of each regional image; Step 4: Using the markers of each regional image, fuse the four regional images to obtain an enhanced image; In step 2, the specific method of converting the four polarization images into the second component TQ image and the third component TU image of Stokes is: Using the formula: TQ(i)=T1(i)-T3(i), TU(i)=T2(i)-T4(i), Calculate and obtain the Stokes second component TQ image and the third component TU image, where i is a pixel index, TQ(i) is the pixel value of the i-th pixel of the second component TQ image, TU(i) is the pixel value of the i-th pixel of the third component TU image, T1(i) is the pixel value of the i-th pixel of the horizontal polarization image, T2(i) is the pixel value of the i-th pixel of the 45-degree linear polarization image, T3(i) is the pixel value of the i-th pixel of the vertical polarization image, and T4(i) is the pixel value of the i-th pixel of the 135-degree linear polarization image; In step 3, the specific method of converting the second component TQ image and the third component TU image of Stokes into four regional images is: pass: Four regional images are obtained, where TS1(i) is the i-th pixel image of the first sub-region, TS2(i) is the i-th pixel image of the second sub-region, TS3(i) is the i-th pixel image of the third sub-region, and TS4(i) is the i-th pixel image of the fourth sub-region; |*| is an absolute value operator, and the pixel index with a pixel value equal to 1 in the four regional images is the mark position information; In step 4, the four regional images are fused using the markers of each regional image to obtain an enhanced image. The specific method is: By formula: The four-region image fusion is realized, where TF(i) is the pixel value of the i-th pixel of the fused enhanced image TF.

2. The electromagnetic wave polarization imaging fusion enhancement method according to claim 1, characterized in that: Step 1 also includes a process of denoising the four polarization images.

3. The electromagnetic wave polarization imaging fusion enhancement method according to claim 1, characterized in that: In step 2, the method for obtaining the pixel values ​​of the pixel points in the four polarization images is: converting the four polarization images into two-dimensional matrix forms respectively, and obtaining the pixel value of each pixel point in the polarization image.

4. An electromagnetic wave polarization imaging fusion enhancement system, used to execute the electromagnetic wave polarization imaging fusion enhancement method according to any one of claims 1 to 3, characterized in that: The system comprises: a focusing lens (2), a polarized antenna (3), a radiometer channel (4) and a data acquisition and processing device (5); The focusing lens (2) is used to focus the electromagnetic waves spontaneously radiated or reflected by the target detection scene onto the polarized antenna (3); The polarized antenna (3) and the radiometer channel (4) are connected to form an image observation unit, which is used to collect horizontal polarization images, 45-degree linear polarization images, vertical polarization images and 135-degree linear polarization images of the target detection scene; The data acquisition and processing device (5) converts the four polarization images into a Stokes second component TQ image and a third component TU image; then converts the second component TQ image and the third component TU image into four regional images; and finally fuses the four regional images to obtain an enhanced image.

5. The electromagnetic wave polarization imaging fusion enhancement system according to claim 4 is characterized in that: The data acquisition and processing device (5) comprises a data acquisition device (51), a data processor (52) and a display (53); The data collector (51) collects four polarization images sent by the radiometer channel (4), and sends the four polarization images to the data processor (52); The data processor (52) is used to convert the four polarization images into Stokes second component TQ images and third component TU images after denoising, and then convert the second component TQ images and third component TU images into four regional images; and finally fuse the four regional images to obtain an enhanced image; The display (53) is used to display the enhanced image.

6. The electromagnetic wave polarization imaging fusion enhancement system according to claim 4 or 5, characterized in that: There is one image observation unit, which is arranged on an observation axis. The image observation unit rotates around the observation axis to collect four polarization images in a time-sharing manner.

7. The electromagnetic wave polarization imaging fusion enhancement system according to claim 4 or 5, characterized in that: There are four image observation units, and the four image observation units collect four polarization images simultaneously.

Images