引入行为特征概念的海量图像数据分类法研究

Mass Image Data Classification Method Research with Introduction of Concept of Behavioral Traits

研究海量图像数据的准确分类问题。图像数据中,图像需要表达的信息过多,形成海量特征,无法找到单一特征对单个图像进行约束。传统的图像数据分类方法通过多特征约束,运用多个特征对图像唯一性进行表达,表达特征过多,影响了分类的效率。为此,提出一种基于行为特征的海量数据分类方法。利用主成分分析方法,对采集的海量图像中的行为特征进行有效的识别,将识别的结果作为图像数据分类的依据,从而缩小图像数据分类的范围。利用人工免疫方法,根据上述行为特征,完成海量图像的数据分类。实验结果表明,利用改进算法进行海量图像数据分类,能够有效提高分类的精确度,从而为不同领域的图像应用提供保障。

Study large amounts of image data accurate classification problem. The paper put forward a classification method for huge amounts of data based on behavioral characteristics. Using principal component analysis method, the behavioral characteristics in huge amounts of image recognition were acquired, and the recognition of the resuits was as a basis for the classification of image data, so as to narrow the scope of the image data classification. Using the artificial immune method and according to the above behavior characteristics, huge amounts of image data were classified. The experimental results show that the algorithm presented in this paper can effectively improve the accuracy of the classification.