Research On Infrared Small Target Detection Algorithm In The Sky Background

Infrared detection systems play an important role in civil and military fields and have become one of the key areas of interest for researchers in various countries.The inherent drawbacks of infrared imaging technology make the infrared image detail texture and contrast low,and the overall visual effect of the image is weak.Due to the long imaging distance,small infrared targets only account for a relatively small number of pixels in the image,and the different environments in which the targets are located,making small targets susceptible to interference by various clutter,making the detection results have a false alarm,which seriously affects the detection performance.In recent years,many researchers have proposed effective small target detection algorithms,but the robustness of the algorithms is not high when the small targets are in different scenes.In this paper,an infrared image enhancement and infrared small target detection method is studied by analyzing the imaging characteristics of small targets and sky backgrounds.The main research includes the following parts:(1)The imaging characteristics of infrared images and small targets against the sky background are analyzed,the basics related to the tensor are studied,the slicing and expansion of the tensor,and the mathematical theorems related to the parametric number are introduced.The theoretical basis is provided for the subsequent infrared small target detection.(2)An infrared image enhancement algorithm based on image layering is proposed.Firstly,the local details of the image are estimated adaptively using an edge-preserving model,and the algorithm enhances from the smoothing layer and the detail layer,using an adaptive Gamma transformation algorithm for the smoothing layer;and an improved spatial entropy enhancement for the detail layer.Finally,the two enhancement results are fused to obtain the final enhanced image.The enhancement experiments are conducted on a variety of infrared databases,and the experimental results show that the enhancement results of the proposed algorithm are visually effective,and the local detail texture is effectively enhanced.(3)An infrared small target detection algorithm based on joint local contrast is proposed.It consists of two modules.First,define a ratio-difference measure to enhance the small target and suppress the background.Second,a constrained difference measure is defined to suppress clutter and enhance the target.The two contrast measures are combined to obtain the saliency map.Finally,an adaptive threshold is calculated to extract the target.Experiments on a series of real IR images and sequences demonstrate that the proposed method can achieve better detection performance than other state-of-the-art methods.(4)An infrared small target detection algorithm based on image tensor decomposition is proposed.First,to better estimate the background component,utilize the Laplace operator to approximate the background tensor rank.Secondly,combined local gradient features and highlighted area indicators to model the local targets prior,which can effectively suppress the complex background clutter.The proposed model was solved by the alternating direction method of multipliers The experimental results on various scenes show that our model achieves an excellent performance in suppressing strong edge clutter and estimating small targets.