加权融合的GLCM的目标物体检索算法

Weighted Fusion of GLCM for Target Object Retrieval

论文提出了一种基于加权融合的灰度共生矩阵(GLCM)的目标物体检索算法。基于GLCM的纹理统计特征的提取方法,分析了GLCM参数对图像特征的影响,提取了0o、45o、90o、135o 4个方向上的灰度共生矩阵特征参数,并对不同方向上特征值进行加权融合,使用最近距离法实现目标物体的快速检索。实验检索准确率可达87.29%。

In this paper, a target detection algorithm based on weighted fusion of gray level co-occurrence matrix（GLCM） is proposed. Based on GLCM, the paper analyzes the influence of GLCM parameters on image features, extracts the characteristic parameters of gray level co-occurrence matrix in four directions, and obtains the optimal weight of eigenvalues in different directions.And the fast and accurate judgment of the target object is achieved by using the nearest distance method. The experimental results can reach 87.29% accurate matching rate. Analysis and experiments show that the proposed algorithm has better recognition rate and robustness to different target objects.