The classification process is an important task in many application of computer image analysis for classifying images based on color or texture low-level features. In this work, a texture classification system is presented which supports querying with respect to texture low-level feature. The fundamental idea is to generate automatically image description by analyzing the image content. The underlying techniques are based on the Gray-Level Co-occurrence Matrix (GLCM) and Gray-Level Run Length Matrix (GLRLM) as statistical approaches to texture analysis. These two techniques are applied in separated manner. Each class is represented by features vector(s) in the features space and stored in a file. Then, a selection to the best set of features is done using fuzzy concepts (triangular membership functions or trapezoidal membership functions). Given the query image, the system first extracts its features vector, and then compares the selected features with those stored in the file to find the nearest class using fuzzy concept. During the evaluation process, it was found that; the best results are obtained from combination among features which in turn achieve higher selection rate for features as well as for whole system (gets selection rate nearly 90% for focombination among four features and 60% without combination).