1 简介
基于计算机视觉实现物体检测。
2 部分代码
function [ background_frame ] = BackgroundExt( input_video )% This script make a background frame based on the samples of the frames. That is, it takes the average of the% frames and save the result as the background frame% If you have the background frame, then you can use it directly. We are% using averaging because we do not have access to the background image,% and by averaging we can supress the effect of moving objects in the% resulting frame and have a better estimation of background frame.sample_step = 2;vid = VideoReader(input_video);%vid = VideoReader('sample.mov');%frame = vid.read(inf);Height = vid.height;Width = vid.width;nframes = vid.NumberOfFrames; %Number of framesbackground = zeros(Height,Width,3); %Initial Background Image%% First Stage: averaging over all of the background samplesfor i=1:sample_step:nframes-sample_stepbackground = background + double(read(vid,i));endbackground = sample_step*background/(nframes);%imshow(uint8(background));background_frame = background;end
3 仿真结果


正在上传…重新上传取消

4 参考文献
[1]邹华宇, 王剑. 一种基于计算机视觉的目标检测方法及系统:.
博主简介:擅长智能优化算法、神经网络预测、信号处理、元胞自动机、图像处理、路径规划、无人机等多种领域的Matlab仿真,相关matlab代码问题可私信交流。
部分理论引用网络文献,若有侵权联系博主删除。
本文介绍了如何使用计算机视觉技术实现物体检测,通过背景帧的平均计算,抑制动态物体对背景的影响,从而提高背景估计的准确性。作者分享了BackgroundExt函数的代码片段,并展示了实际的仿真结果,讨论了邹华宇和王剑的研究方法。
1514

被折叠的 条评论
为什么被折叠?



