Texture feature extraction for endoscopic gastritis images

Abstract

Gastritis is the disease that cause patient to feel uncomfortable in the abdomen. Normally, image on the abdomen will be taken to identify the abnormalities that occurs in the stomach. This report will presents a visualization technique by using a computer where it will apply the characteristics of image to endoscopic gastritis images classification. The main aim was to get a breakdown of characteristics depending on the texture images. Grey-level Co-occurence Matrix (GLCM) features are extracted using Discrete Wavelet Transform (DWT). There were two stage of DWT is applied to the gastritis image. The texture feature were collected for classification process. The endoscopic images are classified into normal and abnormal. The combination of features lead to higher classification rate