Research On Neutron Image Denoising Based On Perona-Malik

Neutron imaging technology is an important non-destructive testing technology,which can visually display the internal structure of the measured object through the image without damaging the measured object.It is widely used in various fields such as industrial testing and medical inspection with its unique advantages in resolution and penetration.In a word,the small neutron imaging system with low fluence rate has good development prospects based on the advantages of convenient use,high cost performance,and good development prospects.However,the imaging system will inevitably be disturbed by various factors during the neutron imaging process,such as neutron scattering,irradiation intensity,dark current,electronic noise,imaging environment,etc.,resulting in noise pollution of the resulting neutron image.It also produces quality degradation such as blurring and reduced contrast.Among them,additive white gaussian noise is the most common type of noise.These degradation phenomena reduce the effectiveness and efficiency of testing and affect the follow-up work.Therefore,research on the denoising method of neutron images can not only obtain clearer internal details from the denoised images,but also reduce the requirements for hardware equipment and the restrictions on the development of neutron imaging equipment.Since there is no standard image for small neutron imaging systems,no reference image quality evaluation index is introduced in order to judge the quality of neutron images scientifically and intuitively.In order to remove gaussian noise in the neutron image and smooth the noise while preserving the details of the image,this paper proposes a denoising algorithm based on the P-M model.In this paper,a new diffusion function with better diffusion effect than the P-M model is constructed by improving the second-order partial differential equations.Combined with the adaptive edge threshold and stopping criterion,this paper established a new denoising algorithm model,which obtains better denoising effect.The experimental results show that this method can effectively remove gaussian noise compared with the traditional P-M method and other partial differential methods,and achieved satisfactory results visually.In the tests with the reference image quality evaluation method PSNR and the non-reference image evaluation methods NIQE and NRQA,this method showed good denoising results while reducing the running time.It provides an important evaluation method for the development of miniaturization of neutron imaging equipment.