Multiple repeated patterns in checked fabrics have partial distribution in the spatial frequency domains. Each distribution around local values approximates the normal distribution. Therefore, distributions similar to the original distribution can be reconstructed using characteristic values of each distribution around local values. Data on the vertical and horizontal axes in the spacial frequency domains permit reconstruction of images similar to the original image. A study reconstructed distributions on the vertical and horizontal axes in the spatial frequency domains to compress checked images and used the Gram-Chrlie approximate equation to reconstruct the distributions. This compression method accounted for local mean, variance, skewness, and kurtosis values only as compression codes. 7 refs.