To determine the location of an optical flow vector, you can calculate the optical flow between two consecutive frames of a video or two images. This can be achieved using the “estimateFlow” function, which is part of the Computer Vision Toolbox. For further information, refer to the documentation using the following command in the MATLAB command line:
web(fullfile(docroot, "/vision/ref/opticalflowhs.estimateflow.html"))
Below is a basic code example for obtaining the location of optical flow vectors:
% Read two consecutive frames
frame1 = imread('Frame1.png');
frame2 = imread('Frame2.png');
% Convert to grayscale
grayFrame1 = rgb2gray(frame1);
grayFrame2 = rgb2gray(frame2);
grayFrame2 = imresize(grayFrame2, size(grayFrame1));
% Initialize optical flow object
opticFlow = opticalFlowHS();
% Calculate optical flow for the first frame
flow1 = estimateFlow(opticFlow, grayFrame1);
% Calculate optical flow for the second frame
flow2 = estimateFlow(opticFlow, grayFrame2);
% Display the second frame
imshow(frame2);
hold on;
% Plot the flow vectors
plot(flow2, 'DecimationFactor', [5 5], 'ScaleFactor', 10);
hold off;
Executing this code will display the image with the optical flow vectors as desired.
