基于图像处理的木粉目数检测与分析

为了解决生产中木粉目数传统检测方法存在的问题，结合先进的数字图像处理技术，在分析木粉颗粒形态特征的基础上，提出了一种基于形态学边缘检测和最大Feret直径的目数检测方法。该方法首先进行图像HIS颜色空间转换，基于S分量进行目标提取，应用多尺度形态学边缘检测算子提取边缘，然后根据颗粒形态特征只保留Feret最大方向上的直径，最后通过单位换算实现目数检测。实验结果表明，该方法具有较高的检测精度，可在生产中推广应用。

Detection and Analysis of Wood Powder Mesh Based on Image Processing

In order to solve the problems with the traditional wood powder mesh detection methods, based on advanced digital image processing technology and analysis of morphological features of wood powder particles, a mesh detection method based on edge detection of morphology and maximum Feret diameter is put forward. In this method, image HIS color space conversion is conducted first, a target is extracted based on S component, edge extraction is conducted using a multi-scale morphological edge detection operator, only the diameter in the direction with maximum Feret is retained according to particle morphological features and finally mesh detection is realized through unit conversion. The particle-size analysis experimental result shows that this method features high detection accuracy and suitable for promotion in production.