Font Size: a A A

Research Of Ground Motion Metal Target Recognition Technology Based On Millimeter-Wave

Posted on:2017-06-29Degree:MasterType:Thesis
Country:ChinaCandidate:Z Z DangFull Text:PDF
GTID:2322330488468486Subject:Navigation, guidance and control
Abstract/Summary:PDF Full Text Request
With the rapid development of millimeter wave technology,millimeter-wave guidance technology has been used gradually on the modern precision-guided weapons to detect the metal target on the ground in combination with the traditional guidance technology.It can significantly improve the performance of the precision-guided weapon.Technology of millimeter wave radiometer passive Detection and recognition is an important research direction.Section 1 presents the background,purpose and significance of the research.The present development situation of the millimeter wave detection and identification technology is also proposed in the first part.Section 2 discusses the relevant theories of the millimeter wave radiometer passive detection of the motion metal target on the ground.Then,the mathematical model of the projectile-target intersection point is built and the simulation results are anyalsied.Compared with the traditional detection model,the range of detection on the ground is wider,and it can monitor the sky.The feature information can be extracted in order to prepare for the target recognition system.The identification method on millimeter wave passive detection and recognition system are propsed in section 3.These methods include fuzzy recognition,BP neural network recognition and RBF network recognition.The target template of the feature information is proposed according to the detection model in section 4.The characteristics of the metallic object information are compared with the feature template in order to use the maximum membership degree principle to achieve the recognition of metal object.Then,the BP neural network is built and trained it to recognise the metal objects.It can identify and analyze the recognition performance,points out its inherent defects.To overcome the defects of BP neural network,using RBF networks to recognise target.Finally,the target recognition performance of three different methods is presented that RNF network is the best.Conclusion is dranw in last section.
Keywords/Search Tags:Feature extraction, Target identification, Fuzzy recognition, BP neural network, RBF network
PDF Full Text Request
Related items