digital image processing

27
Digital IMAGE PROCESSING Amir Hossain 13-23814-1

Upload: ovisopto-amir

Post on 16-Jul-2015

79 views

Category:

Technology


0 download

TRANSCRIPT

Digital IMAGE PROCESSING

Amir Hossain

13-23814-1

What is a image???

What is a Digital Image???

Digital Image

DIGITAL IMAGES are electronic snapshots taken of a scene or scanned from documents, such as photographs, manuscripts, printed texts, and artwork. The digital image is sampled and mapped as a grid of dots or picture elements (pixels).

Our vs Computer vison

Let’s go to the image root(Continue)

pixel

• A pixel (abbr. for picture element) is the smallest unit ofan image.

Let’s go to the image root(Continue)

• MATLAB (Matrix Laboratory) stores images asmatrices.

• In MATLAB, image pixels are referenced using (row,

col) values.

• Origin of the coordinate system (1,1) is the top leftcorner of the image

Let’s go to the image root

0.204.102 00000000 11001100 1100110

Image are two kinds:• Grayscale/Black-white image(16 bit image)Such as 11001100 1100110

Black white• RGB/Color image(24 bit image)Such as 00000000 11001100 1100110

Red(R) Green(G) Blue(B)

• Therefore, a 640x480 image is a matrix of 640 columns and 480 rows, each element of this matrix is called an image pixel.

In a summary

• Digital Image is nothing but collection of pixels which are arrange in matrix form

• Pixel is consists of sub pixels(black and white sub pixels for grayscale and red, blue and green sub pixels for color image)

• This sub pixels have value which represented in 8 bits binary number

What is Digital Image processing?

• Digital image processing is the is a method to convert an digital Image in order to get an enhanced image or to extract some useful information from it by using computer algorithms

Some Applications of image processing

Smoothing Image(Gaussian blur method)

1/91/9

1/91/9

1/91/9

1/91/9

1/9

Origin x

y Image f (x, y)

e = 1/9*106 + 1/9*104 + 1/9*100 + 1/9*108 + 1/9*99 + 1/9*98 + 1/9*95 + 1/9*90 + 1/9*85

= 98.3333

FilterSimple 3*3

Neighbourhood106

104

99

95

100 108

98

90 85

1/91/9

1/9

1/91/9

1/9

1/91/9

1/9

3*3 Smoothing

Filter

104 100 108

99 106 98

95 90 85

Original Image

Pixels

*

The above is repeated for every pixel in the

original image to generate the smoothed image

Image Smoothing Example

Ima

ge

s ta

ke

n fro

m G

on

za

lez &

Wood

s, D

igita

l Im

ag

e P

roce

ssin

g (

20

02

)

Normal Image

Image Smoothing Example

Ima

ge

s ta

ke

n fro

m G

on

za

lez &

Wood

s, D

igita

l Im

ag

e P

roce

ssin

g (

20

02

)

Using 3*3 filter

Image Smoothing Example

Ima

ge

s ta

ke

n fro

m G

on

za

lez &

Wood

s, D

igita

l Im

ag

e P

roce

ssin

g (

20

02

)

Using 5*5 filter

Image Smoothing Example

Ima

ge

s ta

ke

n fro

m G

on

za

lez &

Wood

s, D

igita

l Im

ag

e P

roce

ssin

g (

20

02

)

Using 9*9 filter

Image Compression(Delta Encoding)

Process of delta encoding

• Instate of every pixel value we consider group of pixels where nearby pixels are most similar

• We give value according to similarity which

called delta value

• If neighbor pixels are identical then delta value=0

Image Compression(Delta Encoding)

Continue

• If almost identical then close to 0

• In a high regulation image neighbor pixels are a lot more identical so the delta value looks like this

• If we consider this

• We can express the group of pixel like this

• Png image follow this Encoding which is lossless compression

Viola Jones face detection

Step 1:Haar feature(collect raw face identifier)

Step 2: Integral Image (specific value for identifier)

Integral Image(continue)

Step 3:Adaboost(find necessary information)

Step 4:Cascading(detect face by matching information)

Summary

Detected Image using Viola Jones

Thank YouAny Questions

???