Study On The Enhancement Methods Of Low-contrast Nuclear Imaging

Nuclear imaging technology is a new imaging technology which combines traditional nuclear science technology with modern digital imaging technology.With the help of the transmission capabilities of nuclear radiation and particles associated with nuclear reactions,the nuclear imaging forms the image to show the internal structural information of the measured objects.This technique has been widely used in real applications,such as medical diagnosis,industrial nondestructive testing,life sciences,national defense,and security inspection.The low contrast is a common phenomenon in nuclear imaging,which is caused by the imaging environment,the equipment and other factors,affects the efficiency of nuclear imaging in non-destructive testing.The processing methods of nuclear imaging based on digital image processing technology play an important role in nuclear imaging preprocessing.Specific to these problems,the processing methods of nuclear image edge enhancement and extraction,and denoising and restoration are unfolded.The innovative research work of this thesis includes:1.Edge Extraction and Enhancement method of Nuclear Imaging(1)In order to obtain and enhance the edge structure containing most of the information of the image,an edge extraction and enhancement method based on the sparsity of nuclear imaging data is proposed.Firstly,a counting operation is adopted to filter out the effect of isolated noise points and smooth area pixels on the results.Then,the sparse structure of each image block in the image is described according to the sparse characteristics of the imaging data.Finally,a global sparse representation structure of the imaging data is built,and the selection of threshold value can be solved by combining with the shrinking operations.This method can improve the efficiency of edge enhancement and extraction to some extent.(2)Specific to the problems of the diversity and complexity of degraded nuclear imaging,an edge extraction and enhancement method based on the generalized normed form for nuclear imaging is proposed.This method analyzes the representation of image sparse structure with different norms and compares the effects of different norms structures on edge extraction results and processing time.This method fully considers the sparse structure of nuclear imaging data and removes the pre-filtering smoothing operation,thereby further shortening the time of edge enhancement.2.Denoising and Enhancement method of Nuclear ImagingA digital nuclear imaging enhancement method based on neutron imaging technology is proposed.In the process of neutron imaging,the ?-rays that mixed in the neutron beam can interfere with and damage the CCD component in the imaging system,resulting in some high-brightness,variable-size,randomized distributed white spots occur in the imaging results,called ? spots.These ? spots bring about great difficulties for nuclear imaging analysis and processing.To solve this problem,this paper proposes a denoising and enhancement method for neutron images based on the idea of “detection-corrosion-enhancement”.Firstly,the “detection” the area of gamma spots by adopting the cooperative operation of the spatial adaptive filter and the median filter.Then,“corrode” the area where gamma spots located at by using the traditional median filter.Finally,gamma spots is gradually weakened and eliminated iteratively,and “enhancement” results are output.The digital image processing methods are adopted to improve the quality of nuclear imaging,enhance and restore the important information in the image,improve the contrast of imaging,which has certain guiding significance for the improvement of the accuracy and processing speed of medical diagnosis and industrial nondestructive testing.Practical experiments show that the proposed edge extraction and enhancement method has the advantages of effective extraction and enhancement effects and short processing time.These methods can effectively extract and enhance the edges of nuclear imaging.For the specific ? spots in the neutron images,the proposed method can effectively protect the edges and details of the image while effectively removing the ? spots.At the same time,this method also has stable performance and good results when dealing with other types of noise.