Information Processing Technology Of Omni-directional And Rapid Infrared Search And Track System

Infrared Search and Track(IRST)systems search for,track,and identify potential threats including missiles and airplanes at long-range by distinguishing the differences of infrared radiation emitted by targets and complex background environments,and thus play a passive defense role in defensive early-warning systems.IRST systems have immense advantages,such as wide field,simple structure,good stealthiness,all-time capability and advanced anti-jamming.Through the cooperation between IRST systems and Radar search systems which are based on active detection,complementary advantages can be realized to obtain wide-field and high-performance search and warning systems.Thus,IRST systems have been widely used in sea,land and air military platforms with high value,which generate huge military effectiveness.IRST systems have been studied since the early 1970 s.In the 1990 s,some equipment that meets the technical requirements was developed,such as VAMPIR system of France.It was widely used in western naval surveillance.Infrared linear array device with Time Delay Integration(TDI)work pattern was utilized to scan the omni-directional scene and acquire optical information.The sampling time of scanning is 9~13s and its sampling rate is 0.08~0.1Hz.Since the 21 th century,the missile technology has developed rapidly and its speed has increased from the subsonic speed to several Mach.Besides,its combat modes have been diversified,such as single,multiple and multi-direction attack.Thus,in order to ensure the safety of high-value military platforms including warships,aircrafts,missile launching vehicles,the military power in contemporary world are racing to develop fast IRST systems based on staring imaging with Infrared Focal Plane Array(IRFPA)and image stitching technology(its sampling rate is above 10Hz).The author participated in the research project of omni-directional and fast IRST system in our university,made an intensive study of signal processing and other corresponding technology,and proposed a set of high-performance information and signal processing algorithm applicable to omni-directional IRST system.The main content and innovation points of this paper should be as follows:(1)Preprocessing technology of omni-directional images: the performance of omni-directional and fast IRST system greatly depends on the quality of acquired images.Generally,acquired infrared images have features of high background,low contrast and much noise.Besides,data type of acquired images is 14 bit.Thus,before omni-directional image stitching and target detection,it is necessary to adopt preprocessing to compress image data and enhance its quality.Traditional methods generally ignore the differences between background regions and target regions and enhance both regions simultaneously,which amplifies background noise and decreases performance of omni-directional and fast IRST system.According to this,based on the analysis of current algorithms,an effective image preprocessing approach based on interest region selection is proposed.Interest regions with rich details are detected and detection results are used to compress and enhance interest regions adaptively.Experimental results illustrate that the proposed method can not only reduce processing regions,but also improve contrast of salient regions,which creates favorable conditions for the following conductions of fast image stitching and target detection.(2)Omni-directional image stitching: in high-performance omni-directional and fast IRST systems,omni-directional images are stitched by multi adjacent images in order.Thus,image stitching technology directly affects the quality of omni-directional images and the performance of system.Tradition infrared image stitching methods are generally transplanted from those of visible image.Thus,they are time-consuming and difficult to apply in omni-directional and fast IRST systems.According to this,a fast stitching method is proposed.Coarse registration in Fourier transform domain and fine registration by image feature are utilized to reduce computation in the algorithm.Simulation results show that the proposed method can effectively reduce consuming time while ensuring the stitching accuracy,so it possesses certain practical engineering value in omni-directional and fast IRST system.(3)Dim and small infrared target detection in omni-directional images: dim and small infrared target detection in complex backgrounds is a crucial problem of omni-directional and fast IRST system.There are problems in traditional detection methods,such as high computation or complexity,poor adaptability and so on.According to this,a fast target detection method based on multi-feature fusion is proposed.And a coarse-to-fine detection strategy is used to decrease consuming time.Experimental results indicates that the proposed method can not only detect dim and small infrared targets with small amount of computation,high detection probability,and low false alarm rate,but also adapt to various complex backgrounds.(4)Hardware platform design: to implement algorithm of image and signal processing in omni-directional and fast IRST system,a high speed and parallel processing hardware platform with FPGA and DSP at the core is designed.The experimental results show that the platform can completely achieve various functions and satisfy the real-time requirement.Besides,this platform is easy to extend and with high stability.