The Optimization Design And Application Research Of Matched Filter Based On Wavelet Transform

The two keys of developing the identification performance in optical correlators are integrated structure and excellent stability of recognition,our laboratory have proposed a integrated optical correlator model called planar integrated 2f optical correlator,in this system we have used two phase-only digital micro-lenses instead of the traditional fourier lens,so as the volume has been compressed and the system has been integrated,while at the same time reducing the effective recognition of the system,and the OTSDF matched filter is widely applied to the filed of optics correlators recognition because of its good recognition performance in distortion invariant recognition.In order to improve the recognition performance of planar integrated 2f optical correlators,this paper proposed a new filter algorithm that using wavelet transform to improve the OTSDF filter(W_OTSDF),the multi-scale and band-pass filter characteristics of wavelet transform are used to improve the performance of the planar integrated 2f optical correlators.The simulation verified the relevant output peak intensity and signal to noise ratio(SNR)are improved,the actual optical path is also built to load the matched filter on the spatial light modulator(SLM),and the distortion identification of the targets are achieved.The main research work of this paper is as follows:Firstly,the theoretical of wavelet transform and matched filters are analyzed,proposing an optimization design of OTSDF filter with wavelet transform,and obtaining a matching filtering algorithm which are suitable planar integrated 2f optical correlators,and can improve the recognition performance for distortion target.Secondly,based on the theory of planar optical transfer function and the structural parameters of the planar integrated 2f optical correlators system,the structural parameters of the simulation system are determined.The recognition process of the system is simulated by fresnel diffraction and phase modulation of the lens,The MATLAB simulation model of the system is established,too.Thirdly,the W_OTSDF matched filter algorithm is applied to the simulation model of the system.The effects of the matched filter are simulated and analyzed.The results show that comparing with the original OTSDF,the optimized W_OTSDF filter performance well in peak to correlation energy(PCE),peak to side-lobe ratio(PSR),and SNR when identifying distortion targets.Identify the W_OTSDF matched filters effect better than OTSDF.And the mechanism of the W_OTSDF filter's performance improving is analyzed.Finally,on the experimental platform,the W_OTSDF matched filter is loaded on the SLM to verify the recognition effect of the distortion target.The experimental results show that the system can effectively identify distorted targets within a rotation angle of-48 °~ +44 °,a scaling of 60%~148%,or a rotation angle of 72% ~134% within a scaling range of-32°~+36°.In this paper,the optimization design W_OTSDF filter is better than the OTSDF filter in the peak to correlation energy(PCE),peak to side-lobe ratio(PSR),and SNR,so the W_OTSDF filter has better recognition effect,which is beneficial to improve the recognition ability of the integrated correlators and this algorithm can be widely applied to the planar integrated optical correlator system,the W_OTSDF filter with better ability of distortion invariant recognition can better identify the target and has great application value.