摘要
皮棉异性纤维杂质检测技术是近几年来国内外研究的难点。为有效检测皮棉中与棉纤维外观极其相似的异性纤维杂质,提出了一种显微近红外成像方法用于检测皮棉中异性纤维。该方法将棉纤维与异性纤维在特定红外波段的吸收特性差别,转化为近红外光谱成像系统中两者的灰度、形态图像特征差别,通过显微光路对图像特征差别放大,利用图像分割技术将异性纤维目标分割出来。试验结果表明,采用显微近红外成像方法捕获的图像中,异性纤维灰度、形态特征明显,其检测结果与实际相符,此方法可有效识别皮棉中异性纤维杂质。
Technologies currently used cannot effectively detect foreign fibers in cotton for their almost same appearance as the cotton fibers. In this paper, the method for detecting foreign fibers using micrographic NIR imaging was proposed. According to the method, the absorption characteristic discriminations between cotton fibers and foreign fibers at the near infrared band were transformed to image features discriminations. Then these discriminations were amplified through the micrographic lens on CCD. An image segmentation algorithm was selected for extracting foreign fiber objects from cotton background in image. The result indicated that the features of foreign fibers in the near infrared image were obvious and the conclusion was consistent with the fact. And it provided a feasible and an effective method to detect foreign fibers in cotton.
出处
《农业工程学报》
EI
CAS
CSCD
北大核心
2004年第3期104-108,共5页
Transactions of the Chinese Society of Agricultural Engineering
关键词
异性纤维
红外吸收特性
显微近红外成像
图像处理
foreign fibers
infrared absorption characteristics
micrographic NIR imaging
image processing
