基于YCbCr颜色空间的木材纹理检测技术

Wood Texture Figure Extraction Algorithm Based on YCbCr Color Space

木材的纹理特征是木材缺陷检测和木材材种鉴定中的重要技术,传统的木材纹理检测技术主要集中在提取纹理的边缘,不能获取完整的纹理数据信息,丢失了木材纹理中最重要的颜色特性。针对这一情况,提出了一种基于YCbCr的木纹纹理检测技术,该算法首先对RGB彩色图像进行gamma修正,然后把RGB图像转换为YCbCr图像,在Y-Cb-Cr三维空间中利用分割平面,把纹理和底色区分开来。实验结果表明,该方法能够很好地提取红木类木材的纹理特征,提取的纹理特征连续,提取率高。

Wood texture feature is an important technology in wood defect detection and wood species identification.The traditional wood texture detection technology mainly focuses on the edge of texture extraction,which can not obtain complete texture data information and lose the most important color characteristics in wood texture.In view of this situation,a wood texture detection technology based on YCbCr is proposed.Firstly,the RGB color image is gamma modified,and then the RGB image is converted into YCbCr image.In the Y-Cb-Cr three-dimensional space,the segmentation plane is used to distinguish the texture from the background color.The experimental results show that this method can extract the texture features of mahogany wood very well,the texture features extracted are continuous and the extraction rate is high.