Research On Lung CT Image Processing Method Based On Improved Top Hat Algorithm

As a common lung disease,pulmonary nodule is easy to involve other organs,and even develop into lung cancer,which seriously threatens human health.At present,the chest computed tomography(CT)imaging technology is used to obtain the image of pulmonary nodules in clinic,and the doctors diagnose according to the CT image of the lung,which has become the main diagnostic method.However,there are subjective factors in doctors’ judgment,and image acquisition and processing are also important factors related to the accuracy of diagnosis.How to obtain higher quality CT image of lung is an important problem in medicine.Therefore,this paper has carried on the research,this paper has carried on the following research work.In view of the lung CT image processing,starting from the mathematical morphology and gray image morphology,the characteristics of lung CT image and the process of lung CT image processing are mastered,and the related algorithms of lung CT image processing are studied.Select the top hat algorithm,through the use of top hat algorithm for image processing,image edge extraction,noise reduction and contrast enhancement,further optimize the image quality,easy to identify the characteristic value of pulmonary nodules,calculate the area,perimeter and diameter of pulmonary nodules image,provide reference for clinical diagnosis.In order to improve the effect of image processing,On the basis of top hat algorithm,multi-scale concept and weight concept are fused,and multi-scale i and weight concept are combined λi is added to the algorithm to make it more accurate,At the same time,wavelet processing is introduced to fuse with multi-scale top hat algorithm to form an improved top hat algorithm.The improved top hat algorithm is selected to test the lung CT image processing,and two typical lung nodule images are selected for processing.After multi-step extraction,segmentation and filling,the chain code technology is used to count the eigenvalues of suspected points,and the eigenvalues of lung nodule images are obtained as the basis for clinical diagnosis.In addition,in order to further verify the effect of the improved top hat algorithm,it was compared with the regional growth factor,and six groups of typical CT images of single pulmonary nodule were selected for comparative analysis.The research shows that the improved top hat operator is more effective in medical lung CT image processing,and can obtain complete,accurate and clear gray image,which can provide more accurate diagnosis method for clinical medicine.