Research On Image Enhancement Based On Array Stochastic Resonance

The existence of noise makes the signal seriously affected in the process of processing.Most of the classical filtering methods extract weak signal by removing noise,but the signal is lost or damaged.Stochastic resonance(SR)theory overcomes the shortcomings of classical filtering methods by using its characteristic of converting noise energy into signal energy.It is proved that there is a good side of noise,which is more suitable for weak signal detection in strong noise environment.In order to further improve the effect of weak image signal enhancement and system processing efficiency,this work extends from single stochastic resonance model to array stochastic resonance model,and carries out signal enhancement processing of gray image and binary image.This study has carried out the following work:(1)The dynamic saturated nonlinear system is applied from the field of one-dimensional discrete signal to the field of two-dimensional image signal enhancement,and the research of enhanced processing of different dimension signals is realized.The parallel array saturated stochastic resonance system is studied.The performance of weak BPAM signal is improved by adjusting the parameters of the system and the size of the array unit.When the dynamic saturated nonlinear system processes the noisy signal,the performance of the system is effectively improved by the method of modulation and demodulation.(2)A gray image enhancement method based on array stochastic resonance is proposed.In this method,the two-dimensional image signal is transformed into bipolar aperiodic BPAM input signal suitable for array stochastic resonance model processing by using Hilbert dimension reduction scanning and modulation method.The system output signal is converted into binary signal through the demodulation process,and finally into two-dimensional gray image signal.In the strong noise environment,the simulation results show that the proposed method has better image enhancement effect,higher restoration image quality,and the larger the array unit,the better the image restoration effect.(3)A binary image enhancement method based on stochastic resonance is proposed.The array model of binary image enhancement is established.In this model,the saturated nonlinear system is used as the array processing unit,and the image enhancement effect is effectively improved by dimension reduction,modulation and demodulation.A single saturated nonlinear system is used as the contrast method of image enhancement effect.In this work,several binary images with different sizes and different noise types are enhanced.The simulation results show that this model can effectively improve the peak signal-to-noise ratio of the restored image.The image enhancement algorithm based on array stochastic resonance proposed in this study is innovative in system model design and image processing.The simulation results show that the array stochastic resonance model based on dynamic saturation nonlinear system has good processing effect on both noise gray image and noise binary image,and can significantly improve the gray level sense and image quality of restored image.