A Technological Approach For Image Segmentation

Abstract

Tea is a refreshing drink worldwide and is one of Bangladesh's most significant beverages. The tea leaves are small to medium in size and oval to elliptical in shape with an average length of 5 to 10 centimetres. The development of dark green leaves is in an alternating pattern and is leathery in texture. Diseases such as brown blight and algal leaf can affect tea production, which in turn causes a tremendous loss to farmers and growers. Early identification may help us prevent the disease from spreading to other plants, and may also help to increase production. The research’s main aim is to disease in tea leaf, and to help the farmer grow better tea plant. The method of detection and identification is achieved by means of image processing. The image processing consists of following steps namely segmentation using expectation and maximization, extraction of attributes using statistical and glcm feature extraction techniques, and classification using ANFIS.