Firearm identification based on FIR system characterizing rotation invariant feature of cartridge case image

We propose firearm identification method based on FIR system characterizing rotation invariant feature of cartridge case image. The rotation invariant feature is represented by the absolute value of Fourier coefficients of polar image on circles with different radii of the cartridge casepsilas primer. Then the FIR(Finite impulse response) system characterizing the rotation invariant feature of cartridge casepsilas primer image is introduced. The absolute value of Fourier coefficients obtained from the polar image on the circles with different radii are used as the input and output of the FIR system, respectively. The obtained impulse response of the FIR system is considered as the unique feature of the individual gun. Finally, a firearm can be identified by using the Fisherpsilas linear discriminant function of the obtained impulse response of the FIR system. The firearm identification experiments are conducted to show the effectiveness of the proposed method.