基于粒子群算法的织物组织结构识别

Recognition of Fabric Organization Based on Particle Swarm Optimization Algorithm

提出了一种基于粒子群算法的织物组织结构识别新方法。该方法采用USB数码显微镜摄取布样图像,经直方图均衡化以增加灰图图像对比度、二值化、去除噪声等一系列图像预处理,用经线纬线的宽度法来提取织物组织结构的特征,用粒子群算法进行识别分类。实验结果表明,通过该方法对织物组织结构的识别具有较高的准确率。

A new recognition algorithm is proposed for recognition of fabric organization based on particle swarm algorithm. The digital images of solid woven fabrics is captured by a USB digital microscope. Firstly, histogram equalization is usded to increase the contrast of gray map images, binary and conducted a series of image pre - processing image denoising. Then, width method with warp weft is used to extract the characteristics of the organizational structure of the fabric. Lastly,particle swarm optimization is used to identify categories. Experimental results show that the method has a high accuracy on the recognition of fabric organization structure.