Research And Implementation Of Image Denoising Algorithm For Portable B-type Ultrasonic Diagnostic Equipment

B-type ultrasonic imaging technology plays an important role in auxiliary medical diagnosis.Traditional b-type ultrasonic diagnosis equipment is bulky,expensive and limited to clinical treatment,which is not conducive to accident diagnosis and family medical treatment.With the development of embedded hardware devices,portable b-type ultrasonic diagnostic equipment become possible.However,as the volume of portable equipment is greatly reduced,the image quality displayed by the equipment is poor,and there are many problems such as background noise and electromagnetic interference noise,which affect the image quality.This paper mainly studies the images generated by the portable b-type ultrasonic diagnostic equipment which is developed by the research group,and improves the imaging effect of the portable b-type ultrasonic diagnostic equipment through simulation and realization.The research contents of this paper are as follows:Firstly,the traditional ultrasonic image denoising algorithms are implemented on portable b-type ultrasonic diagnostic equipment,including: ultrasonic signal preprocessing,line smoothing,line filtering,line enhancement and frame correlation.After observing the realized display image,it is found that the denoising effect is poor,and the far field are interfered by electromagnetic noise.By analyzing the spatial distribution characteristics of far-field electromagnetic interference,this paper gives the method to remove the electromagnetic noise interference of b-type image,and carries out the MATLAB simulations and equipment implementations to verify the effectiveness of the denoising method.Secondly,in order to remove the background noise existing in portable b-type ultrasonic images,this paper studies several ultrasonic image denoising methods including the spatial domain of denoising methods such as bilateral filtering and non-local mean filtering,the transform domain of denoising methods such as wavelet threshold denoising and wavelet correlation denoising and the diffusion theory such as anisotropic diffusion filtering and optimization search denoising based on pixel difference index.This paper simulates the above six algorithms,analyzes the subjective and objective parameters of denoising images,compares the longitudinal denoising effects.Thirdly,in order to verify the effect of the denoising algorithm,this paper implements two denoising algorithms which have a better simulation results.This two denoising algorithms are anisotropic diffusion filtering and optimization search denoising based on pixel difference index.After observing the imaging effect of the algorithm,we find that the two denoising methods have different degrees of denoising effect,but the execution time is long,which cannot meet the demand of real-time display of portable b-type ultrasonic diagnostic equipment.Finally,in order to accelerate the execution speed of the denoising algorithm in portable b-type ultrasonic diagnostic equipment,we choose parallel computing to accelerate the image processing algorithm,and realized three denoising algorithms including bilateral filtering,anisotropic diffusion filtering and optimization search denoising based on pixel difference index.Through implementation and observation,it is found that bilateral filtering and optimization search denoising algorithm based on pixel difference index have significantly improved image quality and operation speed,which verifies that the two algorithms have good practicability for the imaging of portable b-type ultrasonic diagnostic equipment.