Design And Realization Of Infrared Weak Target Singularity Detection And Processing Platform

With the development of infrared technology,infrared imaging equipment has a wide range of applications in various fields because of its advantages of long working distance,strong concealment,working day and night,and so on.But in the long range imaging,the target is dim,the detection environment is diverse,and the signal to clutter ratio is very low,which results in the difficulty of target detection.In addition,with the improvement of target mobility and the modernization of weapon system,people put forward higher requirements on the integrity,real-time and stability of target detection system.The traditional target detection method is not ideal in detecting the target under the condition of low signal-to-noise ratio.Combining the time domain information and the airspace information as much as possible will be the development trend of dim target detection at low signal-to-noise ratio.Therefore,it is of great theoretical and practical significance to study the infrared weak target detection method under low SNR condition and design a suitable hardware processing platform for infrared detection system.The contents and achievements of this paper include:(1)Based on the singularity characteristics of the target in the space-time domain,the singularity detection model of the infrared sequence image is constructed.On this basis,the short-time Fourier transform is applied to the singularity detection of infrared dim targets.This method mainly uses the two-level threshold detection method to remove the influence of noise and cloud clutter in the sequence image,and detect the small moving target based on the different characteristics of the short-time Fourier transform time-frequency spectrum of the target,noise and cloud clutter.(2)The hardware platform of infrared weak target detection system based on FPGA is researched and designed.The hardware system uses Xilinx Inc's Virtex6 series XC6VLX240T-1FFG1156 FPGA chip as the core to design,PCIe bus interface as the data interface to achieve high-speed DMA data transmission,high-capacity DDR3 SDRAM as Cache,which provides a hardware verification platform for algorithm implementation.(3)The logic design of each functional module of the hardware platform is completed on the FPGA,and the target detection algorithm proposed in this paper is transplanted on thehardware platform.The whole hardware system consists of image input and output module,image storage module and image algorithm processing module.The host computer sends the frame data of the sequence image through the image input and output module.The board reads the data in the host memory in DMA transmission mode through the PCIe bus interface and stores it in the image storage module DDR3 SDRAM large capacity cache,and the image algorithm processing module reads data in the DDR3 SDRAM to deal with.After the end of processing,the target point information will be stored in the image storage module and the image transfer module writes the processing result to the host memory through the PCIe bus interface,then the host computer reads the target point information and display the target point trajectory.Through the verification of the measured data,the hardware system can effectively validate the detection algorithm proposed in this paper,and detect the weak moving target in the cloud clutter background,which is of practical value in engineering.