论文部分内容阅读
传统视频监控主要是回放原始视频,难以兼顾到夜间模式的监控,现有夜视监控系统其核心技术是依靠红外技术或像增强技术来改善夜间图像质量。在高性能DSP平台上建立了视频采集与显示系统,嵌入Zadeh-X变换算法来实现视频图像的底层图像挖掘,并且根据信息熵判据来自动寻求最佳挖掘参数θ和δ以达到实时挖掘的目的。试验结果表明,视觉效果显著提高,增强了夜间实时监控中视频图像的质量。
Traditional video surveillance is mainly to play back the original video, it is difficult to take into account the night mode of monitoring, the existing night vision monitoring system, the core technology is to rely on infrared technology or enhanced technology to improve night-time image quality. Based on the high performance DSP platform, a video capture and display system is set up. The Zadeh-X transform algorithm is embedded to realize the bottom image mining of video images. And the optimal mining parameters θ and δ are automatically sought according to the information entropy criterion to achieve real-time mining purpose. The test results show that the visual effect is significantly improved, which enhances the quality of video images in real-time monitoring at night.