摘要
为实现对纸张质量的自动化检测,提高纸病识别效率,针对原始纸张图像实施滤波去噪处理,在提高纸张图像质量的基础上通过图像分割和边界跟踪两种手段提取纸病目标位置,并从周长、面积、矩形度、圆形度以及灰度等方面的角度提取纸病位置特征,最终实现对纸病类型的识别.为验证上述方法的可行性,研究选取黑斑、划痕、孔洞、边缘裂缝、亮斑等种纸病并对其数字图像进行识别检测,发现该检测方法能够较为准确地提取纸病轮廓,所取得的纸病特征数据可用于核实分类结果的准确性,对纸病类型的划分具有一定的指导意义.
Abstract
In order to realize the automatic detection of paper quality,improve the efficiency of paper defect identification.In this study,filtering and denoising are implemented for the original paper image,and on the basis of improving the quality of the paper image,the target position of the paper defect is extracted by two means of image segmentation and boundary tracking,and the position features of the paper defect are extracted from the perspectives of circumference,area,rectangle,circularity and gray scale,and finally the identification of the type of paper defect is realized.In order to verify the feasibility of the above method,five kinds of paper defects,including black spots,scratches,holes,edge cracks,and bright spots,were selected and their digital images were identified and detected,and it was found that the detection method could accurately extract the outline of the paper defects,and the obtained paper defect feature data could be used to verify the accuracy of the classification results,which had certain guiding significance for the classification of paper defect types.
基金项目
陕西省教师发展研究项目(SJS2022ZY009)
NETL
NSTL
万方数据