摘要
为了提高数字媒体背景下视频图像的特征识别能力,需要进行视频图像尺度空间融合处理,提出基于小波尺度分解的数字媒体背景下视频图像尺度空间融合算法。构建数字媒体背景下视频图像多维特征采样模型,结合特征配准方法进行数字媒体中视频图像的信息融合和自适应分割,提取数字媒体视频图像的边缘轮廓特征量,采用子空间分割方法进行数字媒体视频图像的多维分割和像素信息特征重组,建立数字媒体视频图像的模糊信息融合模型,采用小波尺度分解方法进行数字媒体视频图像的多维尺度分解,根据尺度分解结果进行高维像素谱信息重组,建立数字媒体视频图像的灰度直方图,采用模糊信息融合方法实现数字媒体视频图像的尺度空间融合。仿真结果表明,采用该方法进行数字媒体视频图像尺度空间融合的自适应性较好,视频图像尺度空间融合输出的信噪比较高。
In order to improve the feature recognition ability of video image in digital media background,it is necessary to carry out video image scale space fusion processing,and a video image scale space fusion algorithm based on wavelet scale decomposition is proposed.The multi-dimensional feature sampling model of video image in digital media background is constructed,the information fusion and adaptive segmentation of video image in digital media are carried out by combining feature registration method,the edge outline feature quantity of digital media video image is extracted,the multi-dimensional segmentation of digital media video image and pixel information feature recombination are carried out by subspace segmentation method,and the fuzzy information fusion model of digital media video image is established.The wavelet scale decomposition method is used to decompose the digital media video image.According to the scale diversity results,the high dimensional pixel spectrum information is reorganized,the gray histogram of the digital media video image is established,and the fuzzy information fusion method is used to realize the scale space fusion of the digital media video image.The simulation results show that the proposed method has good adaptability to scale space fusion of digital media video images,and the signal-to-noise ratio of video image scale space fusion output is high.
作者
周倩
ZHOU Qian(ChangSha Commerce&Tourism College,Hunan Changsha 410116,China)
出处
《新一代信息技术》
2019年第20期88-93,共6页
New Generation of Information Technology
基金
湖南省技术创新引导计划科普专项(项目编号:2017ZK3170)。
