基于RGB颜色特征的海报图像自动检索方法

Automatic Retrieval Method of Poster Image in Based on RGB Color Feature

为了快速、有效检索大量海报图像中特征,提出一种基于RGB颜色特征的自动检索方法,使用户可根据颜色在图像中出现的频率和图像各个色彩对比,确定图像的表现形式及特征。首先使用手绘图形、摄影图片、电脑生成图对海报中表现形式特征及RGB颜色空间解析;采用RGB通道分别等量量化法和图像信息熵计算分析颜色表征出现概率;利用颜色特征提取、选取、量化获得相似性度量函数,运用颜色矩代表图像中颜色的分布情况作为检索矢量,得到最终自动检索方法。仿真实验结果表明,所提方法在外界干扰情况下能获得较高检测结果。

In order to quickly and effectively retrieve the image features in massive poster,an automatic retrieval method based on RGB color features is proposed,which enables users to determine the expression forms and characteristics of the image according to the frequency of color appearing in the image and the comparison of each color in the image.First of all,hand-painted graphics,photographic pictures and computer-generated graphics are used to analyze the performance characteristics and RGB color space in poster;the RGB channel is used to quantify the equal quantity method and image information entropy to calculate and analyze the probability of color representation;the similarity measurement function is obtained by extracting,selecting and quantifying the color features,and the color moment is used to represent the distribution of colors in the image Finally,the final automatic retrieval method is obtained.The simulation results show that the proposed method can obtain higher detection results under external interference.