基于逆投影变换的纱筒纱线余量检测算法

Yarn Bobbin Margin Detection Algorithm Based on Inverse Perspective Transformation

摘要: 纱筒纱线余量检测是纺织行业自动化生产过程中的重要一环，针对目前纱线余量检测算法精度低，本文提出一种基于逆投影变换的纱筒纱线余量检测算法。首先提取纱筒内轮廓，对图像滤波，设计曲率均值高斯卷积核对图像卷积后分割提取纱筒内筒轮廓，然后提取纱筒外轮廓，对纱筒正面区域进行粗分割，选择正面区域轮廓上曲率平稳点建立混合贝塞尔曲线拟合模型，拟合得到纱筒正面区域外轮廓。然后对纱筒图像投影效应进行建模分析，选择纱筒外轮廓上的四个点作为变换基准点，求解逆投影变换矩阵，实现纱筒图像的透视矫正，最后根据制定的纱筒纱线余量计算准则计算得到纱筒纱线余量。在自主搭建的检测平台上使用本文方法进行实验，实验结果证明纱线余量的检测精度在8%以内，满足实际生产要求，为纺织产业自动化生产提供一定的依据。

Abstract: Yarn margin detection is an important part of the automated production process in the textile industry. In view of the low accuracy of the current yarn margin detection algorithm, this paper proposes a yarn margin detection algorithm based on inverse perspective transformation. Firstly, the inner contour of the bobbin is extracted and the image is filtered. The curvature mean Gaussian convolution kernel is designed to segment and extract the inner contour of the bobbin after image convolution. Then the outer contour of the bobbin yarn is extracted and the front area of the bobbin is roughly segmented. The curvature stationary point on the contour of the front area is selected to establish a mixed Bezier curve fitting model, and the outer contour of the front area of the bobbin yarn is fitted. Then the perspective effect of bobbin image is modeled and analyzed. Four points on the outer contour of the bobbin are selected as the transformation reference points, and the inverse perspective transformation matrix is solved to realize the perspective correction of the bobbin image. Finally, the yarn margin is calculated according to the established yarn margin calculation criterion. Experiments are carried out using this method on the self-built detection platform. The experimental results show that the detection accuracy of yarn margin is within 8%, which meets the actual production requirements and provides a certain basis for the automatic production of textile industry.