图像融合技术在非织造材料纤维取向中的应用

Application of image fusion technology in nonwovens' fiber orientation

主要探讨图像处理技术测试非织造材料纤维取向的方法。首先，利用多焦面图像融合技术获取超景深条件下材料的清晰显微图像；其次，利用Canny边缘检测算子提取复杂状态下纤维的2条边缘线来表征纤维取向信息；最后，根据边界曲线积分算法计算出融合图像中纤维段的切线角度，经统计获取非织造材料0°～180。°取向分布曲线。采用扫描电镜图像人工取向测试及短隔距力学拉伸试验结果对上述方法测试结果进行验证，结果表明这种新方法能够快速准确地测试非织造材料的纤维取向。

The application of image processing technique to testing fiber orientation in nonwovens was discussed. Firstly, multi-focal image fusion was employed to capture sharp-focused images in the super depth of field. Secondly, canny boundary detector was used to extract fibers＇ two edges under complex conditions to indicate fiber orientation. Finally, fiber segments＇ orientation angle was calculated using boundary integral algorithm and fiber orientation distribution within the range of 0 to 180 degree can be obtained. Results obtained using image fusion was verified by manually depicting fiber orientations in their SEMs and short-span tensile tests. Experimental results demonstrate that the new method can fast and accurately measure fiber orientations in nonwovens.