The system provides a Standard for fabric defect Detection System using Four Point Grading System. This Four point Grading System is an international standard used by Textile Industry. The proposed system uses Deconvolution wiener filter for the purpose to denoise the images. The images are binarized and segmented using thresholding by separating the pixel values based on intensity level which is either one or zero. After that Deconvolution wiener filter is used to diminish the noise in images while attempting to restore the original signal. This filter which works in the frequency domain is used to denoise the images. Taken together the process segmentation and filtration to get the pre-result image. Finally, the morphological operations both erosion and dilation are done for adjusting the neighborhood pixel values. The combination of erosion and dilation operations is used based on shapes. The features are extracted using region props functions. Then the classification is done with accurate results. The images used in the proposed system are from TILDA’s Texture dataset.