• Optics and Precision Engineering
  • Vol. 31, Issue 6, 872 (2023)
Qingzhu LI, Zhining LI*, Zhiyong SHI, and Hongbo FAN
Author Affiliations
  • Shijiazhuang Campus, Army Engineering University of PLA, Shijiazhuang050003, China
  • show less
    DOI: 10.37188/OPE.20233106.0872 Cite this Article
    Qingzhu LI, Zhining LI, Zhiyong SHI, Hongbo FAN. Single heading-line survey of MGTS for magnetic target pattern recognition[J]. Optics and Precision Engineering, 2023, 31(6): 872 Copy Citation Text show less

    Abstract

    The planar grid measurement of the magnetic gradient tensor system (MGTS) is often utilized for magnetic target recognition; however, it is difficult to measure, complicated to analyze, and requires high instrument precision. In this regard, we propose a magnetic target pattern recognition method based on MGTS single heading-line survey. First, the sensitivity of magnetization direction is analyzed for 15 attributes including the components, eigenvalues, and invariants of the magnetic gradient tensor (MGT). The more sensitive attributes are used to identify target postures, and the insensitive ones are for target shapes. Then, the time-domain signal characteristics of the measured quantities are extracted and the category labels are set. Principal component analysis (PCA) is employed to reduce dimensionality, visualize features, and determine the optimal dimension. Finally, the kernel extreme learning machine optimized by the sparrow search algorithm (SSA-KELM) is used to train and test the survey sample data. The pattern recognition of the magnetic target is hence realized. In the simulation, the recognition of 1) different magnetization direction categories of magnetic dipoles and 2) shape categories of geometric bodies such as the sphere, cuboid, and cylinder is 100% accurate. In the experiment, a total of 180 learning routes were measured for three types of magnets and their corresponding postures. Under the training:testing ratio of 6:4, the results of magnet posture and shape recognition were completely accurate.
    G=BxByBz=gxxgxygxzgxygyygyzgxzgyz-gxx-gyy=v1v3v2λ1λ3λ2v1v3v2-1,(1)

    View in Article

    Gm=1db1x-b3xb1y-b3yb1z-b3zb2x-b4xb2y-b4yb2z-b4zb1z-b3zb2z-b4zb3x+b4y-(b1x+b2y)(2)

    View in Article

    I1=λ1λ2+λ2λ3+λ1λ3c=GFI2=detG=λ1λ2λ3Az=gxz2+gyz2+gxx+gyy2u=-λ32-λ1λ2cosθ=λ3-λ32-λ1λ2=λ3uΔT=bx2+by2+bz2-Tg(3)

    View in Article

    s(I,D)=i=minmaxj=minmaxp(I,D)-p(Ic,Dc)i=minmaxj=minmaxp(Ic,Dc)(4)

    View in Article

    y1   y2      yNT=j=1Lβjgwj,bj,x1j=1Lβjgwj,bj,xNT,j=1, 2, ,L(5)

    View in Article

    gw1,b1,x1gw2,b2,x1gwL,bL,x1gw1,b1,x2gw2,b2,x2gwL,bL,x2gw1,b1,xNgw1,b1,xNgwL,bL,xNβ1Tβ2TβLT=t1Tt1TtLTH(N×L)β(L×m)=T(L×m)(6)

    View in Article

    β=H+T=HTH-1HTT =HTHHT-1T(7)

    View in Article

    fLxi=j=1Lβjgwj,bj,xi=j=1Lβjhjxi=hxiβ,(8)

    View in Article

    β=HTIC+HHT-1T(9)

    View in Article

    fLxi=hxiHTIC+HHT-1T=kxi,x1kxi,xNTIC+K-1T.(10)

    View in Article

    X=x1,1x1,2x1,dx2,1x2,2x2,dxn,1xn,2xn,d,FX=f1x1,1x1,2x1,df2x2,1x2,2x2,dfnxn,1xn,2xn,d(11)

    View in Article

    xi,jt+1=xi,jtexp-iα×tm,  if  R2<STxi,jt+qL,          if      R2ST(12)

    View in Article

    xi,jt+1=qexpxworstt-xi,jti2,      if  i>0.5nxperft+1+xi,jt-xperft+1A+L,if  i0.5n(13)

    View in Article

    xi,jt+1=xbestt+βxi,jt-xworstt,   if  fi>fgxbestt+kxi,jt-xworsttfi-fw+ε,if  fi=fg(14)

    View in Article

    Qingzhu LI, Zhining LI, Zhiyong SHI, Hongbo FAN. Single heading-line survey of MGTS for magnetic target pattern recognition[J]. Optics and Precision Engineering, 2023, 31(6): 872
    Download Citation