Texture Features Extraction of Chest HRCT Images Based on Granular Computing

Combining with signs of lung diseases in highresolution computed tomography (HRCT) images, this paper introduced texture feature extraction into HRCT images analysis. By analyses the relationship of texture features and imaging disease signs, selected three kinds relevant texture features, realized texture parameter extraction of any region of interest (ROI) and straight line, calculated area of lung tissue. This paper presented a segmentation algorithm based on the granular computing theory. This algorithm uses the average gray value of ROI to select seed points automatically. According to tolerance relation system, the criteria of growth is improved, segmented the lung tissue of chest HRCT accurately. The results of extensive experiments illustrate that we can extract texture parameter effectively, gain the data needed for diagnosis of lungs disease. It is the more pertinence and practicality than classical texture analysis methods.