Research On Infrared Small Target Detection Method

Compared with the traditional radar detection system,the infrared detection system is less susceptible to electronic interference and has the advantages of small size,and stable detection performance for low-altitude targets.The infrared detection system has a more stable performance in smoke,fog,and night vision scenarios than the visible light detection system.Infrared target detection technology is one of the core technologies in the infrared detection system.In the long-distance detection scene,the image size of the target is too small,and there is no apparent characteristic information such as shape,texture,color,etc.In addition,there is the interference of complex background.All the above factor makes detecting infrared small targets difficult.This paper studies the background suppression and detection methods of infrared small targets to improve the background suppression effect and reduce the false alarm rate in the target detection results.The main research contents are summarized as follows.1.To improve the suppression effect of traditional local contrast detection algorithms on high brightness point noise and highlighted background edge under complex background conditions.This paper proposed a joint small target detection algorithm based on the maximum median filter and local contrast enhancement to deal with the problem.First,the maximum median filter suppressed the complex background and the bright background edges,and the median filter smoothed the remaining point noise with high brightness.An improved local contrast operator with a suppression weighting factor is used to improve the brightness value of weak and small targets after smooth filtering and further suppress the residual background clutter in the filtering result.The experimental results show that,compared with the existing single-frame infrared small target detection algorithms,the algorithm proposed in this paper has a better background suppression effect and dramatically improves the target signal-to-clutter ratio.2.Aiming at the problem that the single-frame infrared small target detection algorithm has a high false alarm rate under a low signal-to-noise ratio,this paper proposed a small target detection that the spatial-temporal relative local contrast measure.The spatial domain uses the relative local contrast method to suppress the complex background;the temporal domain uses the local difference to repress the relatively stable background area in the image and enhance the moving target.On this basis,spatial and temporal domain detection results are multiplied and fused to highlight the target and offset the residual background clutter in the single-frame detection results of the spatial domain,thereby effectively improving the overall performance of the detection algorithm.Through the simulation verification of the data set,the algorithm proposed in this paper has higher detection accuracy and a lower false alarm rate than the existing infrared small target detection algorithms for the fusion of spatial and temporal domains.