基于图像处理的木片与树皮的新识别参数研究

Approaching to the New Identification Parameter on Wood and Bark Based on Image Processing

利用数字图像处理技术对樟子松、柳木和榆木的木片和树皮图像进行分类识别,首先提取木片和树皮图像的均方差比等6个识别参数,分析其最大值和最小值,然后用支持向量机和BP神经网络对这6个识别参数进行识别研究。结果表明,新识别参数——均方差比,无论用支持向量机,还是BP神经网络,其识别率都较高,因此,均方差比可作为木片与树皮识别的新识别参数。为造纸生产中,将树皮和木片分离,提高纸张质量提供依据。

Wood chips and bark from three tree species （pine, willow and elm） were identified based on digital image processing technology. Six characteristic parameters of wood and bark were extracted to get the recognition rate based on their maximum and minimum values. Support vector ratio and BP neural network were adopted to carry out the identification experiment. Higher ratio of recognition was achieved when the mean variance ratio was used whatever in support vector and BP neural network, indicating that the mean variance ratio could be used as a new identification parameter on wood and bark in paper industry.