Detection of Wood Surface Knots Based on Tezture Features
Introduction The value of a piece of timber is directly related to the quality of wood. The quality is determined by the number of dcfects and their distribution. Knots are the most common defects found during wood inspection. They may decrease the strength of the wood or limit its use for some particular purpose. Traditional methods on wood surface knots detection choose color features and seldom consider the texture features, but the texture features are important characteristics of the knots different from the surrounding. By analyzing the difference between normal wood and wood with knots, color and texture features and parameters correlation analysis, mean, variance, angular second moment, contrast, entropy of the wood images features were selected. To obtain the texture parameters, Gray Level Co-occurrence Matrices (GLCM) method was used. The designed wood surface knots detecting system based on texture features consist of the four stages, as shown by fig. I.
Daguo Ma Weizhu Jin
College of Mechanical and Electrical Engineering, Northeast Forestry University, Harbin, China
国际会议
International Symposium on Wood Science and Technology(IAWPS 2008)(国际木材产品学会联合会2008学术研讨会)
哈尔滨
英文
75-76
2008-09-27（万方平台首次上网日期，不代表论文的发表时间）