A novel four-directional thresholding approach for lung computed-tomography images by using similarity-based segmentation technique

摘要: In automated pulmonary nodules extraction and lung disease diagnosis by image processing techniques, segmentation is utilized as a primary the most essential step of tumour analysis. But due to extensive similarity between vessels, bronchus arteries in region low contrast Computed-Tomography (CT) accuracy highly dependent on precision segmentation. Therefore, precise CT has become challenging preprocessing task for every pathological application. this study, novel Four-Directional Thresholding (FDT) technique introduce d. This propounded segments parenchyma images using Similarity-Based Segmentation (SBS). The proposed aims augment thresholding implementing an advanced approach from four different directions which determination pixels’ value being either foreground or background its adjacent pixel’s intensity final decision made based all directions’ results. study importance neighbour pixels with FDT demonstrated effectiveness method been evaluated images. Eventually result compared other precursors corroborates high exactitude technique.