bilinear interpolation for image amplification (MATLAB implementation)

bilinear interpolation for Image amplification first, the algorithm pseudo-code

Second, the realization

1. Experimental platform and data

This algorithm uses MATLAB language implementation, the experimental platform for Windows 8 32-bit operating system, 4GB memory (usable as 2.31GB), matlab2013b.

Data 1: size: 256*256 of the Lena grayscale image, the implementation of the algorithm will be twice times the amplification operation, as shown in Figure 1:

Figure 1 Grayscale image Lena.png

Data 2: color RGB image of size: 670*502, it will be twice times smaller using the algorithm implemented, as shown in Figure 1 below:

Figure 2 color RGB image he.jpeg 2. Experimental program source code

IMBLIZOOM.M File Source

function [ZI] = Imblizoom (I,ZMF)

%----------------------bilinear interpolation to scale matrices or images---------------------------

% Input:

% I: Image file name or matrix (integer value (0~255))

% ZMF: Zoom factor, which is the multiple of scale

% Output:

% scaled image matrix ZI

% Usage:

% ZI = sselmhsic (' Imagefilename ', ZMF)

% ZMF Zoom to image I and display

% Or:

% ZI = Sselmhsic (I,ZMF)

% ZMF to the matrix I zoom and display

%    ...

%-------------------------------------------------------------------

%%%% Authors:zhi Liu

%%%% Xidian University Student

%%%% email:zhiliu.mind@gmail.com

%%%% date:16-12-2013

Percent STEP1 preprocessing of data

If ~exist (' I ', ' var ') | | IsEmpty (I)

Error (' Input image i ' is undefined or empty. ');

End

If ~exist (' zmf ', ' var ') | | IsEmpty (ZMF) | | Numel (ZMF) ~= 1

The error (' displacement vector zmf ' is undefined or is empty or the element in ZMF exceeds 2. ');

End

If Isstr (I)

[I,m] = Imread (I);

End

If ZMF <= 0

The error (' Zoom multiplier ZMF value should be greater than 0. ');

End

Percent STEP2 the size of the new image with the original image and scaling factor, and creates a new image.

[Ih,iw,id] = size (I);

Zih = round (IH*ZMF); % calculates the scaled image height, the most recent rounding

ZIW = round (IW*ZMF); % calculates the scaled image width, the most recent rounding

ZI = zeros (zih,ziw,id); % Create New image

Percent Step3 extension Matrix I edge

IT = zeros (ih+2,iw+2,id);

IT (2:ih+1,2:iw+1,:) = I;

It (1,2:iw+1,:) =i (1,:,:); IT (ih+2,2:iw+1,:) =i (IH,:,:);

It (2:ih+1,1,:) =i (:, 1,:); IT (2:ih+1,iw+2,:) =i (:, IW,:);

It (:) = i (:); it (1,iw+2,:) = I (1,iw,:);

It (ih+2,1,:) = I (ih,1,:); IT (ih+2,iw+2,:) = I (ih,iw,:);

The percent STEP4 is mapped to the original image (II,JJ) by a pixel (ZI,ZJ) of the new image and interpolated.

For ZJ = 1:ziw% on-column image-by-element scanning

For Zi = 1:zih

II = (zi-1)/ZMF; JJ = (zj-1)/ZMF;

i = floor (ii); j = Floor (JJ); % rounding Down

u = ii-i; v = jj-j;

i = i + 1; j = j + 1;

ZI (zi,zj,:) = (1-u) * (1-v) *it (i,j,:) + (1-u) *v*it (i,j+1,:) ...

+ u* (1-v) *it (i+1,j,:) +u*v*it (i+1,j+1,:);

End

End

ZI = Uint8 (ZI);

Percent% display the same-present matrix p as an image

Figure

Imshow (I,M);

Axis on

Title ([' Original image (Size: ', Num2str (IH), ' * ', num2str (IW), ' * ', num2str (ID), ') ');

Figure

Imshow (ZI,M);

Axis on

Title ([' Scaled Image (size: ', Num2str (Zih), ' * ', num2str (ZIW), ' * ', num2str (ID) ', ') ');

End 3. Experimental results 1) Data 1

Enter Imblizoom (' Lena.png ', 2) in the Command window, and the result is as follows in Figure 3 and figure 4 below:

Figure 3 Operation result of Lena.png 1 (original)

Fig. 4 operation result of Lena.png 2 (figure enlarged twice times) 2) Data 2

Figure 5 He.jpeg Run result 1 (original)

Figure 6 He.jpeg Run result 2 (image zoomed out twice times)

Figure 3~ Figure 6 shows that the program is correct and the effect is very good after zooming in and out. For application and other functions, please review the code comments.