How to make feature vector using optical Flow

2 次查看（过去 30 天）

显示更早的评论

Kong 2020-4-4

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/515388-how-to-make-feature-vector-using-optical-flow

编辑： Kong 2020-4-6

在 MATLAB Online 中打开

Hello.

I want to get a feature vector using the optical Flow Farneback function.

https://www.mathworks.com/help/vision/ref/opticalflowfarneback.html

When I used this code, I got a "flow cell".

How can I make a feature vector using this?

Is it a reasonable feature vector? Which one is better as a feature vector?

I want to use this feature for classification.

% use only flow{i}.Vx
optical_vector{i} = [flow{i}.Vx];
% use all flow values
optical_vector{i} = [flow{i}.Vx, flow{i}.Vy, flow{i}.Orientation, flow{i}.Magnitude];

clear all
close all
%// read the video:
list = dir('*.avi')
% loop through the filenames in the list
opticFlow = opticalFlowFarneback
for k = 1:length(list)
    reader = VideoReader(list(k).name);
    vid = {};
    while hasFrame(reader)
        vid{end+1} = readFrame(reader);
    end
    for i=1:25
        fIdx(i) = i; %// do it for frame 1 ~ 60
        frameGray{i} = rgb2gray(vid{fIdx(i)});
        flow{i} = estimateFlow(opticFlow,frameGray{i});
        optical_vector{i} = [flow{i}.Vx];
    end
 
end