基于视觉传达的多维彩色图像模糊区域特征识别方法

Fuzzy region feature recognition of multidimensional color image based on visual communication

为了提高多维彩色图像的检测识别能力,提出一种基于视觉传达的多维彩色图像模糊区域特征识别方法。构建多维彩色图像模糊区域特征匹配模型,采用模板匹配方法进行多维彩色图像的多尺度结构二值模式分割处理,对分割后的彩色图像利用视觉传达方法进行模糊区域特征提取,获得适应于图像识别表达的像素特征集,采用零均值化的图像滤波方法进行多维彩色图像降噪处理,提高多维彩色图像模糊区域特征检测和识别能力,结合图像的旋转不变性和灰度不变性,实现多维彩色图像模糊区域特征识别。仿真结果表明,采用该方法进行多维彩色图像模糊区域特征识别的准确性较高,特征匹配能力较强。

In order to improve the ability of multi-dimensional color image detection and recognition,a multi-dimensional color image fuzzy region feature recognition method based on visual communication is proposed.The fuzzy region feature matching model of multidimensional color image is constructed,and the multi-scale structure binary pattern segmentation of multi-dimensional color image is processed by template matching method.The segmented color image is extracted by visual communication method,and the pixel feature set suitable for image recognition and expression is obtained.The image filtering method based on zero mean value is used for multi-dimensional color image denoising.The feature detection and recognition ability of multi-dimensional color image fuzzy region is improved,and the feature recognition of multi-dimensional color image fuzzy region is realized by combining the rotation invariance and gray invariance of the image.The simulation results show that the proposed method is more accurate and has better feature matching ability in multi-dimensional color image fuzzy region recognition.