Defects detection inspection is an important task since it highly affects quality, reputation and cost in commercialisation of industrial products. Many solutions and algorithms have been proposed for developing automated inspection systems. In this work, we present a fast method of Edge-Preserving image Smoothing (EPS) for fabric defects detection. Its main advantage is to produce high-quality results at a much lower computational cost, which is sought in this field. The latter is a recent global optimization-based method, named iterative least squares (ILS) that will be used to eliminate the influence of background texture and preserve sharpened significant edges on fabric defects images. Subsequently, the processed fabric images are divided to separate defected and defected-free regions by using binarization method. The proposed work is proven to be effective, robust, and presents good results in detection accuracy compared to other works. .