摘要
对多媒体视频图像显示真实性进行优化,能够有效提高监控视频画面的清晰度。对监控视频图像显示的真实性优化,需要模糊化处理视频图像低频子带,对模糊的隶属度函数进行优化,完成多媒体监控视频图像显示的真实性。传统方法将待增强彩色图像进行转换至色彩空间中,增强色彩饱和度的分量,但忽略了求取其隶属度函数,导致对视频图像的真实度优化效果不理想。提出基于最大模糊熵的监控视频图像显示真实性优化方法。提取视频画面信息,建立保持流形结构近邻图,求解视频画面特征值获取重构权值,得出最优投影;对含有亮度信息的低频子带进行模糊化处理,利用遗传算法对模糊化的隶属度函数进行优化,实现对视频图像亮度的调节,完成对多媒体监控视频图像显示的优化。实验结果表明,利用所提方法能有效提高监控视频图像显示的清晰度,且优化后的视频画面色彩不会产生失真。
Optimization of authenticity of multimedia video image display can effectively improve the clarity of monitoring video picture. To optimize the authenticity of surveillance video and image display, we need to blur the low frequency sub-band of video image, optimize the fuzzy membership function, and complete the authenticity of multimedia surveillance video and image display. The traditional method transforms the color image into color space, enhances the color saturation component, but neglect the membership function, which results in nonideal optimization effect of the video image. The optimization method of video image display based on maximum fuzzy entropy is proposed. Video picture information was extracted, the manifold structure of the nearest neighbor graph was established and maintained, eigenvalue for video reconstruction weights was solved, and the optimal projection was obtained. Low frequency sub band containing luminance information was fuzzified, genetic algorithm was used to optimize the membership function. The video image brightness was adjusted to optimize the multimedia monitoring video image display. The experimental results show that the proposed method can effectively improve the clarity of the monitoring video image display, and the optimized video color will not produce distortion.
作者
张慧
梁银双
ZHANG Hui;LIANG Yin-shuang(School of Information Engineering,Zhongzhou University,Zhengzhou Henan 450000,China)
出处
《计算机仿真》
北大核心
2018年第9期199-202,257,共5页
Computer Simulation
基金
河南省科技厅项目(172400410283)
河南省教育科学规划重点课题(2018-JKGHZD-11)
关键词
多媒体
监控视频
图像显示
真实性优化
Multi-Media
Monitoring video
Image display
Optimization of authenticity
