棉花异性纤维动态识别技术研究.docVIP

棉花异性纤维的动态识别技术研究* 郑文秀1，王金星1，刘双喜1，康玉国2，李付堂2 （1 山东农业大学机械与电子工程学院　泰安　271018；2 北京中棉机械成套设备有限公司　北京　100089） 摘　要：针对棉花异性纤维含量检验过程中手工挑拣方法存在的效率低、误差大等实际问题，研制了一种棉花异性纤维分析仪。该仪器利用皮棉开松机对皮棉进行充分开松，形成不间断均匀棉层，利用线扫描相机对流动棉层进行图像采集。首先对棉花异性纤维原始图像进行灰度处理；再采用自适应阈值技术和滤波技术完成棉花异性纤维图像分割和增强处理，得出清晰的二值化图像；然后采用挖空内点法和邻域搜索法进行轮廓提取，得出异性纤维的特征参数；最后以粗糙集理论为基础，完成了异性纤维的分类识别与重量统计。实验结果表明，该仪器的棉层检测速度达到40 m/h，识别精度达到90%。 关键词：异性纤维；图像处理；轮廓提取；粗糙集 中图分类号：TP391.41　　文献标识码：A　　国家标准学科分类代码：460.4030 Research on dynamic identification technology of cotton foreign fibers Zheng Wenxiu1, Liu Shuangxi1, Wang Jinxing1, Kang Yuguo2, Li Futang2 (1 Mechanical and Electronic Engineering College, Shandong Agricultural University, Tai’an 271018, China;2 Beijing-China Cotton Complete Set of Machinery and Equipment Limited Company, Beijing 100089, China）Abstract：Due to artificial selection method has low efficiency and probable error in the course of the inspection of cotton foreign fibers, a kind of cotton foreign fiber analyzer was developed. After fully-smashing by cotton ginning machine, the uninterrupted uniform cotton layer is formed, and then the image of the flow cotton layer is collected by a line scanning camera. First, gray-scale processing is carried out to the original cotton foreign fiber image; in order to obtain clear binary image, adaptive threshold method and filter technique are used to complete the image segmentation and enhancement processing; then hollowed inner point method and neighborhood search method are used to extract the contours in order to obtain the characteristic parameters of the foreign fibers; finally the category identification and weight statistics of foreign fibers are completed based on rough set theory. Experiments prove that the detection speed of this analyzer can achieve 40 m/h and the recognition precision of this analyzer can reach 90%. Key words：foreign fiber; image processing; contour extraction; rough set 1　引　　言 由于我国棉花主要依靠手工采摘，不可避免地混入了异性纤维（简称异纤）。[1]。异纤在皮棉中的含量虽少，但对纺织品的质量影响严重，一旦混入参与纺织不但影响纺纱能力，