Research On Target Recognition For Anti-jamming Towards Infrared Decoys

This subject comes from a ground infrared tracking test equipment,which can automatically track aerial targets and collect infrared images.In the tracking state,the infrared decoy placed by the target will seriously interfere with the normal tracking performance of the device.This topic analyzes the actual situation of the equipment,creatively defines the feature normalized residual,and formulates the correspondence between the feature normalized residual and the target probability.Using D-S evidence theory to fuse multiple features,the target probability of multi-feature fusion is obtained.In order to eliminate the influence of random errors and consider the actual situation of the equipment,the target probability of multi-feature fusion is used as the basic judgment criterion,and an anti-infrared decoy interference process suitable for this equipment is developed.Rely on the existing hardware platform to implement the process and verify it.It has been verified that it meets the processing requirements of infrared images with a single frame data volume of 640×512×16bit and a frame rate of 100 Hz,and meets the design requirements of more than 90% success rate of anti-infrared decoy interference.At the same time,it also provides a new idea for the stable tracking of long-range infrared targets.The main research contents of this article are as follows:(1)This paper introduces the source and significance of this topic as well as the development status at home and abroad.Combined with the actual situation of the equipment,analyze the problems and challenges in the research process.(2)This paper introduces the typical infrared decoy,and analyzes its defense idea,function principle,usage mode and delivery strategy.When the bait is dropped by the target,the relative positional relationship and characteristics between the two in the infrared image are theoretically analyzed.Determine the research direction through analysis.The research direction is to find and use the difference between the target and infrared decoy in the infrared image after the target put the infrared decoy.(3)In order to accurately extract the features of the target and bait in the infrared image,the image noise reduction and image background suppression are completed by infrared image preprocessing;the sky background is completely removed by image segmentation;five features are extracted by feature extraction.(4)The feature normalized residual is defined creatively,and the correspondence between the feature normalized residual and the target probability is formulated.Introduced the D-S evidence theory and analyzed the defects of its combination rules.In order to avoid the impact of defects,the method of obtaining the target probability is improved.Through D-S evidence theory fusion of multiple features,the target probability of multi-feature fusion is obtained.(5)Based on the target probability of multi-feature fusion and considering the actual situation of the equipment,and an anti-infrared decoy interference process suitable for this equipment is developed.Matlab R2019 b is used to simulate the real data of various scenes,and the effectiveness of the process is verified.Rely on the hardware platform to implement the process and verify it.The verification shows that the success rate of the device against infrared decoy interference reaches more than 90%,and can meet the needs of high real-time processing,meeting the design requirements.