Abstract: By using computer image processing and analysis,a method to measure woven fabric density is developed in this study.Firstly,the fabric image is pre-processed by Wiener2 and histogram equalization to make more clear the fabric texture features;Secondly,the pre-processed woven image is decomposed into horizontal and vertical sub-image based on adaptive wavelet;Thirdly,the features are extracted from the sub-images through some treatments like smoothing;Finally,the fabric density can be obtained by analyzing these features of warps and wefts of the fabric.The experiments have proved that this method can exactly measure the density of three kinds of woven fabric with basic structure.In addition,this paper also presents how to construct the adaptive wavelets.