Change Background: Greenscreen matting without the green screen

Table of Contents

1. Introduction
2. Change Background of an image
3. Change Background of a video
4. Interfaces

• Introduction

Human matting is an extremely interesting task where the goal is to find any human in a picture and remove the background from it. It is really hard to achieve due to the complexity of the task, having to find the person or people with the perfect contour. In this post, I review the best techniques used over the years and a novel approach published on November 29th, 2020. Many techniques are using basic computer vision algorithms to achieve this task, such as the GrabCut algorithm, which is extremely fast, but not very precise.

• Change Background of an image

Upload the image

I am going to turning on the webcam and taking one snapshot of that an then i close my snapshot by using and The function opens a camera preview on the device and returns the current frame in MATLAB® as an RGB image. The resolution of the image is specified by the Resolution property of the camera object cam.

  clc
  clear
  close all 
  warning off 
  ca= webcam; 
  e=ca.snapshot;
  imshow(e);

Upload the new background

I am reading an image that we are going to use to replace our background and that I am also showing.

gk=imread('Nature.JPG'); 
figure; 
imshow(gk); 

create mask on our image

we gonna use imresize that returns image gm that has the number of rows and columns specified by the two-element vector [r c] , createMask that returns a binary mask image with pixels inside the ‘e’ set to true and pixels outside the ‘e’ set to false and mkdir that when the folder creates the folder folderName. If folderName exists, MATLAB issues a warning. If the operation is not successful, mkdir throws an error to the Command Window.

[r, c , b]=size(e);
gm=imresize(gk,[r c]);
mkdir=createMask(e);
figure;
imshow(mkdir);

changing the background

we gonna select the new background (nature) and remove only the body part and we gonna use the uint8 which the Variables in MATLAB of data type (class) uint8 are stored as 1-byte (8-bit) unsigned integers.

gs=gm.*uint8(~mkdir);
figure;
imshow(gs);
subplot(1,2,1);
imshow(e);
subplot(1,2,2);
imshow(gs+e.*uint8(mkdir));
gm=gk;

RESULT

we gonna replace our face and our body part on the new image. The Color Thresholder app creates a new tab displaying the image and the color component controls for the HSV color space. In this color space, H represents hue, S represents saturation, and V represents value. The HSV color space uses a dual-direction knob for the H component and two histogram sliders for the S and V components. The tab also contains the point cloud representation of the colors in the image.

clc
clear  
close all
warning off
ca= webcam;
e=ca.snapshot;
imshow(e);
clear ca;
gk=imread('Nature.JPG');
figure;
imshow(gk);
[r, c , b]=size(e);
gm=imresize(gk,[r c]);
mkdir=createMask(e);
figure;
imshow(mkdir);
gs=gm.*uint8(~mkdir);
figure;
imshow(gs);
subplot(1,2,1);
imshow(e);
subplot(1,2,2);
imshow(gs+e.*uint8(mkdir));
gm=gk;


function [BW,maskedRGBImage] = createMask(RGB)
%createMask  Threshold RGB image using auto-generated code from colorThresholder app.
%  [BW,MASKEDRGBIMAGE] = createMask(RGB) thresholds image RGB using
%  auto-generated code from the colorThresholder app. The colorspace and
%  range for each channel of the colorspace were set within the app. The
%  segmentation mask is returned in BW, and a composite of the mask and
%  original RGB images is returned in maskedRGBImage.

% Auto-generated by colorThresholder app on 13-Aug-2020
%------------------------------------------------------


% Convert RGB image to chosen color space
I = rgb2hsv(RGB);

% Define thresholds for channel 1 based on histogram settings
channel1Min = 0.196;
channel1Max = 0.000;

% Define thresholds for channel 2 based on histogram settings
channel2Min = 0.000;
channel2Max = 0.236;

% Define thresholds for channel 3 based on histogram settings
channel3Min = 0.322;
channel3Max = 1.000;

% Create mask based on chosen histogram thresholds
sliderBW = ( (I(:,:,1) >= channel1Min) | (I(:,:,1) <= channel1Max) ) & ...
  (I(:,:,2) >= channel2Min ) & (I(:,:,2) <= channel2Max) & ...
  (I(:,:,3) >= channel3Min ) & (I(:,:,3) <= channel3Max);
BW = sliderBW;

% Invert mask
BW = ~BW;

% Initialize output masked image based on input image.
maskedRGBImage = RGB;

% Set background pixels where BW is false to zero.
maskedRGBImage(repmat(~BW,[1 1 3])) = 0;

end

• Change Background of a video

im gonna using the same code white loop while that iteratively executes its child components while a specified condition is true. The While Loop component must have at least one child component; the purpose of this component is to run its children several times. If it does not have any children, this component does not add anything to the report.

clc
clear 
close all
warning off
ca= webcam;

gk=imread('Nature.JPG');
while true
    e=ca.snapshot;
    [r, c, b]=size(e);
    gm=imresize(gk,[r c]);
    mkdir=createMask(e);
    gs=gm.*uint8(~mkdir);
    subplot(1,2,1);
    imshow(e);
    subplot(1,2,2);
    imshow(gs+e.*uint8(mkdir));
    gm=gk;
end


function [BW,maskedRGBImage] = createMask(RGB)
%createMask  Threshold RGB image using auto-generated code from colorThresholder app.
%  [BW,MASKEDRGBIMAGE] = createMask(RGB) thresholds image RGB using
%  auto-generated code from the colorThresholder app. The colorspace and
%  range for each channel of the colorspace were set within the app. The
%  segmentation mask is returned in BW, and a composite of the mask and
%  original RGB images is returned in maskedRGBImage.

% Auto-generated by colorThresholder app on 13-Aug-2020
%------------------------------------------------------


% Convert RGB image to chosen color space
I = rgb2hsv(RGB);

% Define thresholds for channel 1 based on histogram settings
channel1Min = 0.196;
channel1Max = 0.000;

% Define thresholds for channel 2 based on histogram settings
channel2Min = 0.000;
channel2Max = 0.236;

% Define thresholds for channel 3 based on histogram settings
channel3Min = 0.322;
channel3Max = 1.000;

% Create mask based on chosen histogram thresholds
sliderBW = ( (I(:,:,1) >= channel1Min) | (I(:,:,1) <= channel1Max) ) & ...
    (I(:,:,2) >= channel2Min ) & (I(:,:,2) <= channel2Max) & ...
    (I(:,:,3) >= channel3Min ) & (I(:,:,3) <= channel3Max);
BW = sliderBW;

% Invert mask
BW = ~BW;

% Initialize output masked image based on input image.
maskedRGBImage = RGB;

% Set background pixels where BW is false to zero.
maskedRGBImage(repmat(~BW,[1 1 3])) = 0;

end

• Interfaces

created by : -MZOUDI Khaoula

Project Link:(https://https://github.com/khaoulamzoudi/traitement-dimage/edit/main/README.md)

Encadré par : Dr. Meryam ELMOUHTADI

(back to top)