夜视红外激光社区视频异动监控预警系统设计

Design of community video abnormal monitoring and early warning system for night vision infrared laser

针对传统方法缺少融合滤波处理过程,导致监控预警图像几何特征提取不及时的问题,提出夜视红外激光社区视频异动监控预警模型优化。硬件模块包括传感器模块、时钟终端控制模块、社区监控图像传输模块、视频总线调度模块以及输出控制模块;通过构建红外激光社区视频异动夜视监控图像的信息采集模型,进行红外激光社区视频异动夜视监控图像的自适应特征提取,根据异常特征提取结果,进行监控信息识别;在监控信息识别基础上,依据异动监控的判别函数,对子空间结构块进行匹配,实现红外激光社区视频异动监控的动态特征提取。根据动态场景中监控视频的异常动态特征,得到红外激光社区视频异动监控图像异常动态谱特征分布,利用稀疏特征量和特征分布进行区域像素融合,实现夜视红外激光预警。实验测试结果显示,采用该方法进行红外激光社区视频异动监控预警系统设计,系统的输出稳定性较好,集成控制能力较强。

In order to solve the problem that the traditional method lacks fusion filtering processing process,which leads to untimely extraction of geometric features of monitoring and early warning images,an optimization of night vision infrared laser community video abnormal monitoring early warning model is proposed.The hardware module comprises a sensor module,a clock terminal control module,a community monitoring image transmission module,a video bus scheduling module and an output control module.By constructing the information acquisition model of infrared laser community video abnormal night vision monitoring image,the adaptive feature extraction of infrared laser community video abnormal night vision monitoring image is carried out,and the monitoring information is identified according to abnormal feature extraction results.On the basis of monitoring information identification,the sub-space structure blocks are matched according to discriminant function of the abnormal monitoring,and the dynamic feature extraction of the infrared laser community video abnormal monitoring is realized.According to the abnormal dynamic characteristics of monitoring video in dynamic scene,the abnormal dynamic spectrum feature distribution of infrared laser community video abnormal monitoring image is obtained,and the regional pixel fusion is carried out by using sparse feature quantity and feature distribution to realize infrared laser early warning of night vision.The experimental test results show that the laser infrared community video abnormal monitoring and early warning system is designed by this method.The output stability of the system is good and the integrated control ability is strong.