2-D frame correlation

clear
camera is that reports a video frame (30 frames/second and maximum 300 frames) at instant t
represented by the vector F(x, y, t) = [FR(x, y, t) FG(x, y, t) FB(x, y, t)],
where x and y are the coordinates of the pixel in
the frame, while FR, FG and FB represent the level of RGB
components at this pixel.
The 2-D correlation between the current frame and a previous frame at instant to can be dened
by the vector C as follows:
C=[Cr;Cg;Cb]=[corr(Fr(x,y,t),Fr(x,y,t0));corr(Fg(x,y,t),Fg(x,y,t0)) ;corr(Fb(x,y,t) ;Fb(x,y,t0))]
How can I represent The 2-D correlation between the current frame and a previous frame at instant t0 (modify this below code) ?
all
close all
clc
%step_1
%%Extract frames from video
%% source https://youtu.be/GF6guS2DIWY
a=VideoReader('camera.mp4');
% for img=1:a.NumberOfFrames;
for img=1:300;
filename=strcat('frame',num2str(img),'.jpg');
b=read(a,img);
imwrite(b,filename);
end
%step_2
%%%read the download frames from video%%%%
%quantization each R,G,B equivalently to 8*8*8
NumberOfLevelsForR=8;
NumberOfLevelsForG=8;
NumberOfLevelsForB=8;
N=NumberOfLevelsForR*NumberOfLevelsForG*NumberOfLevelsForB;
%Enter name of folder from where you want to upload pictures with full path
srcFile=dir('C:\Users\M7MD\Desktop\EXTVIDEO\*.jpg');
for i=1:length(srcFile)
filename=strcat('C:\Users\M7MD\Desktop\ExtractFramesFromVideo\',srcFile(i).name);
I=imread(filename);
%split image into R,G & B
R=double(I(:,:,1));
G=double(I(:,:,2));
B=double(I(:,:,3));
[x,y,z]=size(I);
%create final histogram matrix of size 8*8*8
hist=zeros(NumberOfLevelsForR, NumberOfLevelsForG, NumberOfLevelsForB);
%create col vector of indicies for later refernce
rows=x;cols=y;
index=zeros(rows*cols, 3);
%%%(Maximum Level D 255) for each of the R, G and B components
MR=ceil(NumberOfLevelsForR.*R/255);
MG=ceil(NumberOfLevelsForG.*G/255);
MB=ceil(NumberOfLevelsForB.*B/255);
%2-D frame autocorrelation between current and previous
if i>1
corrR(i)=corr2(MR(i),MR(i-1));
corrG(i)=corr2(MG(i),MG(i-1));
corrB(i)=corr2(MB(i),MB(i-1));
end
C=[corrR;corrG;corrB];
end

回答(0 个)

类别

帮助中心File Exchange 中查找有关 Image Segmentation and Analysis 的更多信息

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by