Fabric pattern recognition based on olfactory neural network

ince fabric textures are diversified and processing of their images susceptible to noise jamming,this paper presents a new fabric recognition method based on olfactory neural network(ONN).It first establishes a biological ONN model and regulates the input voltage of the KⅢ model according to the stimulus-response relationship.Then,the collected fabric image is projected along the directions of X and Y,and the size of the fabric intersection points are calculated.The fabric intersection points are rearranged later,and the obtained images are trained by ONN.The experimental results show that ONN has higher recognition rate of plain and stain weaves than BP neural network.In addition,by taking a new extraction model of fabric intersection points,its recognition rate of twill weaves has also improved significantly.