This document discusses various image processing techniques in MATLAB, including: 1. Using imfinfo() to get metadata about an image, impixelinfo to query pixel values, and rgb2ind() to convert images to indexed color for reduced file size. 2. How colormaps like jet and spring can be applied to images. 3. Techniques for creating grayscale images, dealing with data type issues when displaying images, and extracting individual bit-planes from an image.

1. MATLAB & IMAGE PROCESSING 2
Presented by
Dr. M. Venkatanarayana, cri,ksrmce,kadapa

2. Getting image information
After completion of this topic, student will be able to use
● imfinfo()
● impixelinfo
● rgb2ind()
● colormap(jet)

3. The imfinfo() gives us lots of information
>> info=imfinfo('Sun.jpg')
info =

4. Querying a pixel info
● Because the image itself is an array, we can check the values it is storing.
For example, if we want to know the values of the pixel at (350, 250):For example, if we want to know the values of the pixel at (350, 250):
● >> img(350,250,:)
● ans(:,:,1) = 157
● ans(:,:,2) = 70
● ans(:,:,3) = 0

5. rgb2ind()
● "Indexed color saves a lot of memory, storage space, and transmission
time: using truecolor, each pixel needs 24 bits, or 3 bytes. A typicaltime: using truecolor, each pixel needs 24 bits, or 3 bytes. A typical
640x480 VGA resolution truecolor uncompressed image needs 640x480x3
= 921,600 bytes (900 KiB). Limiting the image colors to 256, every pixel
needs only 8 bits, or 1 byte each, so the example image now needs only
640x480x1 = 307,200 bytes (300 KiB), plus 256x3 = 768 additional bytes to
store the palette map in itself (assuming RGB), approx. one third of the
original size."

7. Code:
● img = imread('Sun.jpg');
● [IND,map] =
● img = imread('Sun.jpg');
● [IND,map] = ● [IND,map] =
rgb2ind(img,32);
● figure('Name','Indexed
image with 32 Colors');
● imagesc(IND);
● colormap(map);
● [IND,map] =
rgb2ind(img,32);
● figure('Name','Indexed
image with 32 Colors');
● imshow(IND);
● colormap(map);

9. 200
400
600
200 400 600 800 1000 1200 1400 1600
800
1000
1200

10. Colormap(jet);colormap(spring)
200
400
600
200
400
600
200 400 600 800 1000 1200 1400 1600
600
800
1000
1200
200 400 600 800 1000 1200 1400 1600
600
800
1000
1200

11. Creating grayscale image: We're going to create two
grayscale images of size 320×240,320×240,one with pixels
all black and the other one all white:
h = 240;
w = 320;
white = uint8(255*ones(h,w));
black = uint8(zeros(h,w));
figure;figure;
subplot(121);
imshow(white);
subplot(122);
imshow(black);

12. Class - Data Type: There are one thing to keep in mind when we process
images: data type. Is my image in uint8or double? One of the most frequent
issues caused by the data type is imshow(). We converted the image to double
to do something with the image, and now we want to draw it
● img = imread('cameraman.tif');
● img_d = double(img);● img_d = double(img);
● % ... played with the image, now we want to display it
● imshow(img_d);
Result is white image rather than original image
The imshow() has two function overloading. It takes image with type uint8 as it ranges
between [0, 255], and takes image with type double as it ranges between [0, 1]. So, in our
case, every pixel with over 1 is considered saturated, and that's why we got white cameraman.

13. img = imread('cameraman.tif');
img_d = double(img);
imshow(img_d/255);
Data type conversion problems

14. Extracting bit-plane from Grayscale image With mod() operation, we can extract a bit-
plane image. The mod(img,2) gives us either 0 or 1
img = imread('cameraman.tif');
img = double(img);
bp0 = mod(img,2);
imshow(bp0);
title('bit-plane 0');
bit-plane 0 bit-plane 1
title('bit-plane 0');

15. bit-plane 2 bit-plane 3 bit-plane 4
bit-plane 5 bit-plane 6 bit-plane 7