Example – WebCamera –possible applications: Motion detection, Earthquake monitoring (or Camera vibration) Intruder detection, alarm function Steganography*, finding hidden messages in original source vs. modified version)

• Can be independent of WebCamera – SURF section can be skipped for better performance

-MATLAB change detection, using general or WebCamera pictures - here is Scale Invariant difference of two similar pictures

File : difference.m (for MatLab) is complex code and Sqrt, Arctangent in SURF algorithms to do jpg to rgb absolute differences – this would not be fast on GR-PEACH. SURF is used as its 3-5x faster than SIFT.

Motion detection, Earthquake monitoring (or Camera vibration) Intruder detection, Alarm function Steganography

MATLAB CODE to take 2 jpgs and determine the scale invariant (SURF algorithm used) absolute difference (like xor ) tic; [FileName_f, PathName_f] = uigetfile('*.jpg','Select the fixed image'); fixed = imread([PathName_f, FileName_f]);

[FileName_m, PathName_m] = uigetfile('*.jpg', 'Select the moving image'); moving = imread([PathName_m, FileName_m]);

out_path = uigetdir;

convert jpg to gray scale

fixed_g = rgb2gray(fixed); moving_g = rgb2gray(moving);

movingAdjusted=imadjust(moving_g); fixedAdjusted=imadjust(fixed_g);

perform scale invariant feature transform via faster SURF vs. older SIFT

ptsFixed=detectSURFFeatures(fixedAdjusted);

ptsMoving=detectSURFFeatures(movingAdjusted);

[featuresFixed,validPtsFixed]=extractFeatures(fixedAdjusted,ptsFixed);

[featuresMoving,validPtsMoving]=extractFeatures

indexPairs = matchFeatures(featuresFixed,featuresMoving);

matchedFixed=validPtsFixed(indexPairs(:,1));

matchedMoving=validPtsMoving(indexPairs(:,2));

Get estimates geometric transform [tform,inlierMoving,inlierFixed]=estimateGeometricTransform(... matchedMoving,matchedFixed,'affine');

Tinv=tform.invert.T;

ss=Tinv(2,1);

sc=Tinv(1,1);

recover scale and rotation angle scaleRecovered=sqrt(ss*ss + sc*sc);

thetaRecovered=atan2(ss,sc)*180/pi;

movingRegistered=imwarp(moving,tform,'OutputView',imref2d(size(fixed)));

perform absolute difference matrix calculation - NoN GPU version diff = imabsdiff(fixed, movingRegistered);

imshow(diff)

cd(out_path);

output file = difference.jpg imwrite(diff, 'difference.jpg', 'jpg');

timeSpent=toc;

clc;

Speeded up robust features - Wikipedia, the free encyclopedia https://en.wikipedia.org/wiki/Speeded_up_robust_features

Wikipedia In computer vision, Speeded Up Robust Features (SURF) is a local feature detector and descriptor that can be used for tasks such as object recognition or registration or classification or 3D reconstruction. It is partly inspired by the scale-invariant feature transform (SIFT) descriptor.

How I can integrate matlab in GR-peach?