Design And Implementation Of X-ray Image Quality Automatic Enhancement System

Since Roentgen discovered X-rays and applied them to image human tissue,the research on X-ray machines has never stopped,and people have been trying to use lowdose mode imaging and obtain high-quality X-ray images.Because the images acquired by low-dose mode imaging are characterized by low brightness,insufficient contrast,and excessive noise,it is necessary to use a computer to enhance the low-dose X-ray images to obtain medical images that meet clinical standards.Thesis mainly studies the X-ray image enhancement algorithm under low dose.In order to prove the effectiveness of the algorithm,thesis shoots and collects data from different hospitals and laboratories,and uses more than 600 image sequences with more than 4000 images of different tissue parts to test the algorithm studied in thesis.The photographed parts include the spine,hip bone,palm,hip joint,ankle joint,thorax,head,etc.The improved performance is obtained from two objective evaluation indicators of user’s subjective evaluation and image quality,which proves that the proposed algorithm is superior to other medical image enhancement algorithms.The main research content of thesis is divided into four parts.1)Research and implement an automatic adjustment algorithm for secondary exposure dose based on image features.The algorithm adjusts the dose of the main exposure process according to the pre-exposed image eigenvalues,and the adjustment process is carried out in multiple times,which can prevent the radiation attenuation from affecting the imaging quality.The exposure time of the system is short,and the exposure time of a single frame can be controlled within 60 ms.The system interacts with image quality to ensure that the average grayscale value of the imaged tissue area falls within the standard grayscale range,reducing the equipment waste rate.2)A multi-frame averaging denoising and anti-artifact algorithm based on Kalman filter is researched and implemented.When the imaging object moves violently,the algorithm performs optimal estimation on the continuously exposed X-ray images,and eliminates the artifact information left on the image caused by the movement of the object.Use multi-frame average noise reduction to eliminate inter-frame noise and fuse to output a composite image with low noise and no artifacts.3)Research and implement algorithms for edge detail enhancement in low-dose Xray images.The algorithm further enhances the details and contours of the composite image from three perspectives: noise removal,edge sharpening,and contrast adjustment.The algorithm combines the ideas of guided filtering and pyramid enhancement,which can filter out image noise,enhance the tissue contour of the imaging site on the X-ray image,highlight the effect of internal details of the tissue,and output a clear X-ray image.4)Analyze user requirements and design system operating interface.Based on the VS platform,the image enhancement algorithm is packaged and the dynamic link library is made.Based on.net framework,use c# to design and develop user interface.The operation interface requirements are suitable for non-computer professionals,with simple operation and complete functions as the standard.