Efficient segmentation of the lung carcinoma by adaptive fuzzy-GLCM (AF-GLCM) with deep learning based classification

摘要

Image processing is an innovative method to convert the real image into a sharp digital image by applying various functions upon it. However, it is a difficult task for physicians in the medical field. The significant difficulty is with the segmentation of images due to the blurred contrast and artifacts at the boundary edges. Hence in this paper, an efficient and adaptive fuzzy-GLCM based segmentation method was proposed. The images derive from the process of bronchoscopy. The ultimate goal of the proposed methodology was the accurate recognition of the lung carcinoma, which undergoes segmentation. The adaptive F-GLCM segmentation method enables the early and easy detection of lung cancer, which helps both the physicians and the patients for proper initial medication. Then the classification was done with the help of the GoogLeNet CNN architecture, which will reveal whether the cancerous growth was in a benign or in a malignant stage. Then the performance analysis of the proposed method was measured by comparing it with the other existing methodology.