| Target recognition in a complex background is a research hotspot in the military field,especially for camouflage target recognition.Traditional intensity detection technology can only detect the radiant intensity of light,so when the target color is similar to the background color or the detection environment is weak or there is dense fog,the traditional intensity detection technology cannot meet the actual detection needs.Compared with traditional intensity detection technology,multi-spectral polarization imaging detection technology has the advantage of obtaining polarization information and spectral information of objects that cannot be obtained by traditional detection technology,and different objects exhibit different spectral characteristics and polarization characteristics,and different There is a clear difference between the polarization information and spectral information possessed by an object,so using multi-spectral polarization imaging detection technology helps to identify hidden objects in the background.This paper based on multi-spectral polarization imaging detection to carry out research on the identification method of camouflage targets.First of all,the energy is used to represent the thickness of the texture distribution in the image.First of all,according to the degree of polarization image and the angle of polarization image reflect the texture thickness of different objects,there is energy in the gray image to indicate the uniformity of the gray distribution and the thickness of the texture,and the target to be detected is significantly different from the background texture.Therefore,the two are normalized before fusion based on energy features.The fused image does not make full use of the gray range of the image itself,and does not fully reflect the difference between the target to be detected and the background,so the image after the energy feature fusion is enhanced with an improved histogram equalization to further improve The contrast between the target and the background.However,due to the lack of consideration of the detailed edge information of the scene of the image,resulting in the problem of loss of detailed edge information in the fused image,combined with the rich edge contour and other detailed information of the light intensity image,it is proposed to use NSCT to enhance the enhanced energy feature image and light The intensity image is fused,considering that the high frequency part of the image mainly reflects the details such as the edge texture of the image,and the low frequency part of the image mainly reflects the overall contour information of the image.In this paper,a strategy based on principal component analysis is used for thelow-frequency part,and a fusion strategy is used for the high-frequency part to take the larger edge.Experiments prove that the fusion image can effectively highlight the texture and other information of the artificial camouflage target,and retain the details of the scene.In order to verify the effectiveness of the algorithm,this paper uses the FD-1665 polarization imager and filters at 490 nm,550nm,670 nm,and 850 nm bands to conduct multispectral polarization imaging detection experiments on camouflaged targets under complex backgrounds,and merges the fusion results with the original images.Comparison of contrast,the results show that the method proposed in this paper can more effectively highlight the target to be measured from the complex natural background,and the fused image has several kinds of information entropy,standard deviation,average gradient,contrast between target and background,etc.Commonly used indicators are superior to the original image,indicating that the method has improved quality compared with the original image.Compared with the fusion image of the traditional algorithm,the target-background contrast is greater than the traditional algorithm,indicating that the algorithm proposed in this paper Can effectively improve the ability of target recognition. |
PDF Full Text Request