CN-116310687-B - Image pseudo-color fusion method and system based on polarization angle characteristics under dark background

CN116310687BCN 116310687 BCN116310687 BCN 116310687BCN-116310687-B

Abstract

The invention provides an image pseudo-color fusion method and system under dark background based on polarization angle characteristics, wherein the method comprises the steps of collecting emergent light intensity of a target scene by utilizing a polarization camera, calculating Stokes parameter image data of the target scene according to the emergent light intensity, analyzing the Stokes parameter image data to obtain polarization information, mapping synthesized light intensity I, linear polarization degree P and polarization angle A to brightness V, saturation S and hue H in an HSV model respectively to obtain normalized synthesized light intensity, linear polarization degree stretching data, polarization angle preprocessing data and hue matching parameters, processing characteristic data of the polarization angle A by utilizing preset polarization robustness judging logic aiming at a current observation point in the target scene to obtain polarization robustness judging parameters, judging the state of polarization noise according to the vibration robustness judging parameters, and converting a pseudo-color HSV image into an RGB color representation image for display.

Inventors

WU LINGXIA
XUE MOGEN
WU YUNZHI
HAN YUSHENG
WANG YONG
HUANG QINCHAO
Zu hongyu
WANG SHUO
HUANG QIHENG

Assignees

中国人民解放军陆军炮兵防空兵学院

Dates

Publication Date: 20260505
Application Date: 20230321

Claims (4)

1. The image pseudo-color fusion method under the dark background based on the polarization angle characteristic is characterized by comprising the following steps of: S1, collecting the emergent light intensity of a target scene in at least 3 directions by using a polarization camera Calculating Stokes parameter image data of the target scene by using preset logic, and analyzing to obtain polarization information, wherein the polarization information comprises linear polarization degree And polarization angle ; Collecting the outgoing light intensity of a target scene by using a polarization camera Processing to obtain the Stokes parameter image data, and collecting emergent light intensities of a target scene in three directions by using a polarization camera 、、 Substituting the formula: to obtain the combined light intensity Linear polarization parameter 、 ; The linear polarization degree is obtained by analysis by the following logic The polarization angle : (4) (5); S2, combining the synthesized light intensity I and the linear polarization degree The polarization angle Respectively mapping to brightness V, saturation S and hue H in an HSV model to obtain pseudo-color model mapping application data, wherein the pseudo-color model mapping application data comprises normalized synthesized light intensity, linear polarization degree stretching data, polarization angle preprocessing data and hue matching parameters; S3, processing the polarization angle by utilizing preset polarization robustness discrimination logic To obtain a polarization robustness discrimination parameter; performing polarization robustness discrimination by using the following logic to obtain the polarization robustness discrimination parameters : (13) (14) (15) Wherein, the , , Corresponding to one pixel point in the image f 1 , M is the number of pixel points in the neighborhood S, wherein the neighborhood S takes eight neighborhood, M=9, then Representation pair As a template 3 3, Neighborhood average, g is the polarization robustness characterization parameter of the current point, t is a discrimination threshold, close and open are morphological open and close operations, and B is a structural element; S4, judging the state of polarization noise according to the polarization robustness judging parameter, taking the maximum value of the normalized synthesized light intensity and the linear polarization degree stretching data to map to the brightness V when the polarization noise is smaller than a preset polarization noise threshold value, and when the polarization noise is larger than or equal to the preset polarization noise threshold value, taking the normalized synthesized light intensity Mapping to the brightness V to obtain a pseudo-color HSV image; Pseudo-color model mapping is performed using the following logic: (10) (11) (12) S5, converting the pseudo-color HSV color representation into an RGB color representation image for display so as to display the pseudo-color image on a computer.
2. The method for pseudo-color fusion of images in dark background based on polarization angle characteristics according to claim 1, wherein said step S2 uses the following logic to compare said linear polarization degree Stretching to obtain the linear polarization degree Stretching to a range of applicable value of linear polarization degree to obtain the linear polarization degree stretching data : (7)。
3. The method for pseudo-color fusion of images in dark background based on polarization angle characteristics according to claim 1, wherein in step S2, the polarization angle is processed by the following logic To obtain the polarization angle pre-processing data: (8); normalizing the emitted light intensity by using the following logic to transform the value range of the emitted light intensity into an applicable light intensity interval so as to obtain the normalized synthesized light intensity : (6); The polarization angle is calculated by the following logic Multiplying a preset coefficient to match the value range of the tone H, thereby obtaining the tone matching parameter phi: (9)。
4. a polarization angle characteristic based image pseudo-color fusion system for performing the polarization angle characteristic based image pseudo-color fusion method in the dark background according to any one of the preceding claims 1 to 3, characterized in that the system comprises: The polarization information analysis module is used for collecting the emergent light intensity of the target scene in at least 3 directions by using the polarization camera Calculating Stokes parameter image data of the target scene by using preset logic, and analyzing to obtain polarization information, wherein the polarization information comprises linear polarization degree And polarization angle ; A data preprocessing module for preprocessing the composite light intensity I and the linear polarization degree The polarization angle The brightness V, the saturation S and the tone H in the HSV model are respectively mapped to obtain pseudo-color model mapping application data, wherein the pseudo-color model mapping application data comprises normalized synthesized light intensity, linear polarization degree stretching data, polarization angle preprocessing data and tone matching parameters, and the data preprocessing module is connected with the polarization information analysis module; The polarization robustness judging module is used for processing the polarization angle by utilizing preset polarization robustness judging logic Acquiring a polarization robustness judging parameter, wherein the polarization robustness judging module is connected with the data preprocessing module; The pseudo-color model mapping module is used for judging the state of polarization noise according to the polarization robustness judging parameter, taking the maximum value of the normalized synthesized light intensity and the linear polarization degree stretching data to map to the brightness V when the polarization noise is smaller than a preset polarization noise threshold value, and taking the normalized synthesized light intensity when the polarization noise is larger than or equal to the preset polarization noise threshold value Mapping to brightness V to obtain a pseudo-color HSV image, wherein the pseudo-color HSV image is connected with the data preprocessing module and the polarization robustness judging module; And the pseudo-color display module is used for converting the pseudo-color HSV color representation into an RGB color representation so as to display a pseudo-color image on a computer, and is connected with the pseudo-color model mapping module.

Description

Image pseudo-color fusion method and system based on polarization angle characteristics under dark background Technical Field The invention relates to the technical field of target identification and imaging, in particular to a method and a system for fusing pseudo-colors of images under dark background based on polarization angle characteristics. Background The contrast of the target is low in a dark background, detection and identification are not facilitated, and the polarized imaging can acquire the target intensity and the polarized information at the same time, so that the method has detection advantages under the conditions of low illumination and the like. The color recognition capability of human eyes is greatly better than the gray level recognition capability, so researchers put forward a plurality of image pseudo-color fusion methods based on polarization information, thereby improving the target detection recognition efficiency. The most classical pseudo-color fusion method is a Wolff based HSV model mapping method, for example, a polarized image visualization method based on color image fusion is disclosed in the prior patent application publication No. CN111292279A, which firstly converts a polarized image into a three-channel image of light intensity, polarization degree and polarization angle, secondly decomposes the image into a polarized color image and a natural light gray image, thirdly performs image fusion on the decomposed two images, specifically selects a proper edge-preserving fusion algorithm according to requirements to fuse light intensity channels of the two images, then normalizes the fused light intensity and polarization angle channels, and finally projects the normalized image into an HSV color space to obtain a final polarized color image with visualization. The aforementioned prior art maps polarization angle to hue, polarization to saturation, and composite light intensity to brightness. However, if the irradiance of the target image is low, that is, the synthesized light intensity component is low, the color characteristic of the fusion result is not obvious enough, the target enhancement effect is poor, and the influence of noise is large. In order to enhance the pseudo-color fusion effect, a pseudo-color fusion method based on non-negative matrix factorization and HSV space is proposed by Zhou et al, a pseudo-color image fusion method based on low-rank sparse factorization is proposed by Xu et al, and the like, which have a certain effect on suppressing background noise, but do not solve the problem from the polarization characteristic only from the viewpoint of information processing. In summary, the prior art has the technical problem that the false color fusion result is not ideal due to the excessively dark background area of the target. Disclosure of Invention The invention aims to solve the technical problem of unsatisfactory pseudo-color fusion results caused by too dark background areas where targets are located in the prior art. The invention adopts the following technical scheme to solve the technical problems that the image pseudo-color fusion method under the dark background based on the polarization angle characteristic comprises the following steps: S1, collecting emergent light intensity I (theta) of a target scene in at least 3 directions by using a polarization camera, calculating Stokes parameter image data of the target scene by using preset logic, and analyzing to obtain polarization information, wherein the polarization information comprises linear polarization degree P and polarization angle A; S2, mapping the synthesized light intensity I, the linear polarization degree P and the polarization angle A to brightness V, saturation S and hue H in an HSV model respectively to obtain pseudo-color model mapping application data, wherein the pseudo-color model mapping application data comprises normalized synthesized light intensity, linear polarization degree stretching data, polarization angle preprocessing data and hue matching parameters; S3, processing characteristic data of the polarization angle A by utilizing a preset polarization robustness judging logic to acquire a polarization robustness judging parameter; s4, judging the state of the polarized noise according to the vibration robustness judging parameter, taking the maximum value of the normalized synthesized light intensity and the linear polarization degree stretching data to map to the brightness V when the polarized noise is smaller than a preset polarized noise threshold value, and when the polarized noise is larger than or equal to the preset polarized noise threshold value, taking the normalized synthesized light intensity Mapping to brightness V to obtain a pseudo-color HSV image; s5, converting the pseudo-color HSV color representation into an RGB color representation so as to display a pseudo-color image on a computer. In the pretreatment process of P and A, the invention fully c