In order to achieve real-time detection of pattern fabric defects in the process of weaving, a machine vision-based rapid real-time detection method is proposed. In off-time case, through the distance matching function of flawless image and the analysis of extreme weight to obtain the periods of pattern fabric accurately, extract the standard element, establish a sequence of the element, extract features and construct a fuzzy classifier. In real-time case, extract features of every image block based on the sequence established in off-time case and then classify the blocks. Construct the accurate classifier if flaw might exist. Experiments shows that, the method put forward in this paper can realize real-time detection for most class of fabrics in different patterns with only one-time learning from the flawless images. The method has a high instantaneity that an average time of detection is about 200 ms and a low fallout ratio that the accuracy rate is above 99%.