贾伟, 王正勇, 张杰, 李伟. 一种基于改进的CS-LBP算子纹理图像自适应检索方法[J]. 微电子学与计算机, 2013, 30(9): 75-78.
引用本文: 贾伟, 王正勇, 张杰, 李伟. 一种基于改进的CS-LBP算子纹理图像自适应检索方法[J]. 微电子学与计算机, 2013, 30(9): 75-78.
JIA Wei, WANG Zheng-yong, ZHANG Jie, LI Wei. A Self -Adaptive Texture Image Retrieval Method Based on the Improved CS -LBP Operator[J]. Microelectronics & Computer, 2013, 30(9): 75-78.
Citation: JIA Wei, WANG Zheng-yong, ZHANG Jie, LI Wei. A Self -Adaptive Texture Image Retrieval Method Based on the Improved CS -LBP Operator[J]. Microelectronics & Computer, 2013, 30(9): 75-78.

一种基于改进的CS-LBP算子纹理图像自适应检索方法

A Self -Adaptive Texture Image Retrieval Method Based on the Improved CS -LBP Operator

  • 摘要: 具有自适应性的方向局部二值描述符(SD -LBP)通过引入自适应对比性的阈值 t增加对图像局部区域及局部纹理过渡区域信息的处理,使图像检索过程中产生的噪声对检索效果的影响达到最小。采用3组实验方案对SD-LBP算子与相关算子的检索效果进行了比较分析。实验结果显示,本文算子在保持CS-LBP算子数据维数不变的条件下,对图像噪声处理更具鲁棒性,同时在检索效果、视觉特性等方面具有明显的优势。

     

    Abstract: The self -adaptive direction local binary operator (SD-LBP) increases the information processing of the image local region and the transition region of image texture,so that the impact of image noise reaches a minimum value.Three groups of experimental program are used to analyze the retrieval performance of the SD -LBP operator and other operators.The image retrieval results based on texture spectrum show that when the operator keeps the CS-LBP data dimensions unchanged,the SD -LBP operator has stronger robustness in the processing of image noise,and it also has obvious advantages in image retrieval efficiency and visual properties.

     

/

返回文章
返回