Satellite Image Classification by Using Distance Metric Dr. Salem Saleh Ahmed Alamri Dr. Ali Salem Ali Bin-Sama Department of Engineering Geology , Department of Engineering Geology , Oil & Minerals Faculty, Aden University, Oil & Minerals Faculty, Aden University, Aden, Yemen Aden, Yemen Dr. Abdulaziz Saleh Yeslam Bin –Habtoor Department of Electronic and Communication Enigeerinng, Faculty of Engineeninng & Petrolem,Hadramote University, Mokula , Yemen Abstract— This paper attempts to undertake the study satellite image classification by using six distance metric as Bray Curtis Distance Method, Canberra Distance Method, Euclidean Distance Method, Manhattan Distance Method, Square Chi Distance Method, Squared Chord Distance Method and they are compared with one another, So as to choose the best method for satellite image classification. Keyword: Satellite Image, Classification, Texture Image, Distance Metric, I. INTRODUCTION Defines image classification as particular class of pattern recognition. Classifiers are described under board categories such as supervised and unsupervised classifiers, parametric and non-parametric, fuzzy classifier and knowledge base classifiers [1]. Defines three major steps involved in the typical supervised classification procedures as follows: • Training Stage: The analyst identifies reprehensive training area and develops a numerical description of the spectral attributes of each land cover type of interest in the scene. • Classification Stage: Each pixel in the image is categorized into the cover class it most resembles. if the pixel is not matching to any predefined class then it is labeled as unknown. • Accuracy Assessment: The classified image is compared with some reference image or ground truth to check the accuracy of the classification [2]. They are proposed an algorithm for very high-resolution satellite image Classification that combines non-supervised segmentation with a supervised Classification and the result show very good performance of approach in comparison to existing techniques [3].Minimum distance classification method in satellite Image is a simple and quick method that does not include covariance They are proposed an algorithm for very high-resolution satellite image Classification that combines non-supervised segmentation with a supervised Classification and the result show very good performance of approach in comparison to existing techniques [3].Minimum distance classification method in satellite Image is a simple and quick method that does not include covariance information and Maximum likelihood classification method is widely used in remote sensing image and can be regard as one of the most reliable techniques [4].They are proposed a new algorithm for texture classification based on logical operators is presented. Operators constructed from logical building blocks are convolved with texture images. This algorithm is applicable to different types of classification problems which is demonstrated by segmentation of remote sensing images, compressed and reconstructed images and industrial images[5]. He proposed a simple scheme which used local linear transformations and energy computation to extract texture features. This simple scheme often gives good results but is not consistent in performance. The statistical methods share one common weakness, of primarily focusing on the coupling between image pixels on a single scale and are also computationally intensive processes [6].He proposed texture segmentation and classification for texture features images based on the grey level co-occurrence probabilities (GLCP) [7].They is proposed classification method based on the Discrete Cosine Transform (DCT) coefficients of texture images by used two popular soft computing techniques namely neuron computing and neuron-fuzzy computing [8].He is proposed to perform unsupervised image classification based on texture features by using a novel evolutionary clustering International Journal of Computer Science and Information Security (IJCSIS), Vol. 14, No. 3, March 2016 https://dx.doi.org/10.6084/m9.figshare.3153877 65 https://sites.google.com/site/ijcsis/ ISSN 1947-5500