基于遥控侦察仪的目标检测

Target detection based on remote control reconnaissance instrument

针对原有的遥控侦察仪只能够采集和显示视频,而未对视频做相应处理的问题,设计了一个基于遥控侦察仪的目标检测方案,赋予遥控侦察仪检测目标的功能。当侦察区域内出现运动目标,遥控侦察仪的显示界面发出警报提示,实现遥控侦察仪的智能化。采用单高斯模型进行背景建模,但检测到的目标图像存在空缺部分。利用三帧差分法得到的差分图像和更新单高斯模型的学习速率来弥补图像的空缺部分,使得检测到的目标图像较为完整。依据帧数来实现学习速率的更新。多次实验表明,遥控侦察仪目标检测性能稳定,满足了遥控侦察仪的功能需求。

Aiming at the problem that the original remote control reconnaissance instrument can only capture and display video without corresponding processing of the video,a target detection scheme based on the remote control reconnaissance instrument is designed to give the remote control reconnaissance instrument the ability to detect the target.When a moving target appears in the reconnaissance area,the display interface of the remote control reconnaissance device issues an alarm prompt to realize the intelligence of the remote control reconnaissance device.This paper uses Single Gaussian Method to build the background model,but there are gaps in the detected target image.The differential image by threeframe differential method is obtainedand the learning rate of Single Gaussian Method is updated to make up for the missing part of the image.This method makes the detected target image more complete.According to the number of frames,the learning rate of Single Gaussian Method is updated.After several experimental tests,the results show that the performance of the image processing system for detecting targets is stable,which meets the functional requirements of remote control reconnaissance instrument.