Research On X-ray Image With High Dynamic Range Denoising And Compression Algorithm

Since the 1970s and 1980s,it has become international practice to carry out security checks on the carry-on belongings of people entering and leaving important places.In recent years,with the continuous improvement of China’s comprehensive national strength,rail transit and aviation have made great progress.The number of passengers who take long-distance buses,high-speed railways and airplanes in China is gradually increasing every year,which also puts forward higher requirements for China’s public security.X-ray detection plays a particularly important role in the field of national public security.It can scan all kinds of packages through the transmission method,and generate translucent images so that security personnel can check whether there are prohibited items in the distance in time.Therefore,it is imperative to improve the imaging ability of X-ray security equipment.In this paper,the characteristics of radio-graphic images are studied,and the actual requirements are combined to improve the image quality,the screening success rate of security personnel and shorten the identification time,so as to meet the relevant requirements of security inspection.This paper mainly includes the following points:Firstly,this paper introduced the research background and significance of the subject,and introduced the domestic and foreign research status of radio-graphic image correlation processing in detail,as well as the development status of security technology.This paper introduced the core components of X-ray safety inspection machine,including X-ray source,detector and collimator.At the same time,the occurrence and detection principle of X ray were introduced in detail,and the evaluation standard of image quality was given.In this paper,several common denoising algorithms were introduced and compared.At the same time,aiming at the problem that X-ray image is easy to produce noise,a filtering method based on neural network and improved block-matched filtering method were proposed to remove noise.Firstly,the PCNN network was used to locate the noise points,and the improved median filter was used to denoise.Then an improved block-matched filter was used to carry out secondary filtering while preserving the edge information of the image.Finally,the experimental simulation was carried out and the result was compared with the traditional algorithm.Several common tone mapping algorithms were introduced.At the same time,an adaptive tone mapping algorithm based on multi-scale local edge preserving filter was proposed according to the characteristics of radio-graphic images.In this algorithm,a local edge-preserving filter was used to filter the high and low energy images at the same time.The detail layer was adaptive enhanced and the base layer was compressed.Finally the low dynamic range image was obtained by merging.Through experiment comparison,it was found that the processing effect and processing time of the algorithm meet the security inspection requirements.