数字图像库中彩色图像信息特征检索优化

Color Image Information Characteristics in Digital Image Library Search Optimization

在对图像库中彩色图像信息特征进行检索优化的研究中,为提高图像处理的效率与质量,在进行信息特征检索优化时,需要对彩色图像信息特征检索向量的组成空间进行准确分析,建立彩色图像数据特征检索模型,但是传统的模糊控制决策算法主要通过对彩色图像信息存储结构进行遍历完成特征检索优化,忽略了彩色图像信息特征检索向量的组成空间,不能建立精确的数据特征检索模型,存在检索不准确、与实际信息特征差异大的问题。提出一种基于大数据分析的图像库中彩色图像信息特征检索优化方法。对彩色图像信息监测特征进行提取与分类,将建立的概念格构造模型与概念格差异融合算法结合,并采用大数据模型和辅助矩阵方法,分析彩色图像信息检索向量组成空间,建立彩色图像信息特征检索模型,并利用相似度系数对模型进行优化,完成彩色图像信息特征检索的改进。仿真结果表明,采用改进的算法进行彩色图像信息特征检索,提高了彩色图像信息特征检索性能,同时降低了图像信息特征检索的差错率。

It needs to analyze vectorial space of color image in digital image information feature searching and build its searching model to increase efficiency and quality of image processing. However,traditional fuzzy controlling decision algorithm optimizes feature searching via traversing the storage structure. The algorithm ignores feature searching vectorial space and cannot build exact data feature searching model. It leads inaccurate searching and great difference with practical information feature. In the paper,we propose an optimization method based on big data analysis. We extract and classify the color image information searching feature. We also combine the Concept Lattices（ CL） tectonic model with the CL difference fusion algorithm. Big data model and assistant matrix method are used to analyze the color image information searching vectorial space. We modify feature searching via building the searching model and optimizing model through similarity coefficient. Simulation results show that the modified algorithm improves color image information feature searching performance and reduces its error rate.