image processing - Universitas Dian Nuswantoro

advertisement
Ferda Ernawan, Ph.D
Postgraduate Program of Dian Nuswantoro University
Email: [email protected]



Course Implementation
Lecture: 2 hrs per week for 10 weeks (Total=20 hrs)
Course Evaluation:
Course Works
Marks
Assignments 1 (Critique a conference article)
20%
Case Study
20%
Research Proposal (5 pages)
30%
Examination
Marks
Final Examination
30%
Total
100%
Lecture 1
Introduction
“Digital Image Processing”, Rafael C. Gonzalez & Richard E.
Woods, Addison-Wesley, 2002
◦ Bahan materi yang di ambil dari buku tersebut
Sub
◦
◦
◦
◦
◦
◦
pokok pembahasan meliputi:
Pengertian digital image?
Pengertian digital image processing?
Contoh digital image processing dalam kehidupan sehari-hari
Aspect of Image Processing
Fundamental step in digital image processing
Matlab as a research tools for image processing and
applications
◦ Characteristic of a good paper in term of image processing and
applications
◦ Penjelasan assignment 1

Sebuah gambar digital adalah representasi dari gambar dua
dimensi sebagai himpunan terhingga dari nilai digital, atau
biasa disebut elemen gambar atau pixel.


Nilai sebuah pixel biasanya merupakan tingkat abu-abu,
warna, tingkat kecerahan atau kekeruhan, dll.
Digitalisasi sebuah image adalah menggambarkan sebuah
kejadian nyata.
1 pixel
Nilai suatu pixel merupakan single number dari
suatu intensitas gambar atau warna gambar.
Digital Image merupakan multidimensional array
dari intensitas gambar atau warna gambar.
99
60

85

32
70 56 78
90 96 67

85 43 92

65 87 99
◦ 1 sample per point (Black & White or Grayscale)
◦ 3 samples per point (Red, Green, and Blue)
◦ 4 samples per point (Red, Green, Blue, and “Alpha”)







GIF (Graphic Interchange Format)
PNG (Portable Network Graphics)
JPEG (Joint Photographic Experts Group)
TIFF (Tagged Image File Format)
PGM (Portable Gray Map)
FITS (Flexible Image Transport System)
BMP (Bitmap)

Digital Image Processing fokus pada dua tugas
utama:
◦ Peningkatan informasi gambar untuk interpretasi manusia
◦ Pengolahan data image untuk penyimpanan, transmisi,
dan representasi untuk autonomous machine perception.

Penggunaan teknik pengolahan citra digital telah meningkat
dan image processing application sekarang telah digunakan
untuk semua jenis pekerjaan pada semua jenis bidang,
misalnya:
◦ Image enhancement dan restoration
◦ Artistic effects
◦ Medical visualisation
◦ Industrial inspection
◦ Law enforcement
◦ Human computer interfaces

Salah satu penggunaan yang paling umum dari teknik Digital
Image Processing yaitu meningkatkan kualitas, remove noise,
dan sebagainya.
Image yang terdegradasi
The result of image restoration

Efek Artistik digunakan untuk membuat gambar lebih
menarik dan untuk menambahkan suatu efek khusus.

Ambil irisan dari scan hati, dan menentukan batas
batas antara jenis jaringan
◦ Gambar dengan abu-abu mewakili tingkat kepadatan
jaringan.
◦ Gunakan filter yang cocok untuk menyorot tepi.
Original Image of a Heart
Edge Detection Image


Manusia sebagai operator memerlukan biaya mahal.
Mesin dapat melakukan pekerjaan lebih cepat, dan
sistem industri seperti itu sudah digunakan dalam
semua jenis industri.

Teknik pemrosesan gambar digunakan secara luas oleh
penegak hukum
◦ Monitoring nomor pelat mobil dalam memonitor
kecepatan / sistem tol otomatis
◦ Pengenalan sidik jari
◦ Peningkatan CCTV gambar

Sistem Informasi Geografis
◦ Digital image processing digunakan secara ekstensif untuk
memanipulasi citra satelit.
Color image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Problem Domain
Knowledge base
Object
Recognition
Color image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Object
Recognition
Problem Domain
Color image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Object
Recognition
Problem Domain
Color image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Object
Recognition
Problem Domain
Colour image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Object
Recognition
Problem Domain
Color image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Object
Recognition
Problem Domain
Color image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Object
Recognition
Problem Domain
Color image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Object
Recognition
Problem Domain
Color image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Object
Recognition
Problem Domain
Color image
processing
Wavelets and
multi resolution
processing
Image
Compression
Morphological
Processing
Image
Restoration
Segmentation
Image
Enhancement
Representation
& Description
Image
Acquisition
Object
Recognition
Problem Domain



The basic data structure in Matlab is the array,
and the operation is sequence, like C
programming.
Matlab stores image as two dimensional array
(matrices),
each element of the matrix
represents a single pixel in the displayed image.
By default, Matlab stores the data in array of
class double.


Image processing toolbox is collection of function of
Matlab numeric computing environment.
toolbox functions implement specialized image
processing algorithm.







Geometric operations
Neighborhood and block operation
Linier filtering and filter design
Transformation domain
Image analysis and enhancement
Binary image operations
Region of interest operation.

The Image Processing toolbox supports basic
type of image, for example:
◦ Indexed images
◦ Binary images
◦ RGB images



In Matlab , the red, green, and blue component of RGB
image reside in single m-by-n-by-3 array.
m and n are the number of rows and columns of pixels
in the image, and the third dimension consists of red,
green, and blue intensity values.
Each pixel in the image, the red, green, and blue
elements are combine to create the pixel’s actual color.
An RGB array can be:
 Class double, in this case it contains values in
the range [0,1]
 Class uint8, in this case the data range is [0,255]



Reading in image data from files, and writing image data out
to file.
Converting images to other image types.
Working with uint8 arrays in Matlab and the Image Processing
toolbox.
You can use the Matlab function to read image data from files.
Imread(‘filename’)
Imread function can read these graphics file formats:
 TIFF (Tagged Image File Format)
 JPEG (Join Photographics Expert’s Group)
 HDF (Hierarchical Data Format)
 BMP (Windows Bitmap)
 XWD (X-Window Dump)



To write image data from Matlab to a file, you can
use imwrite function. Imwrite can write the same
file formats that imread reads.
See the references entire for imread and imwrite for
more information about these function.
In addition, you can use imfinfo function to return
information about the image data in a file.
Function
Purpose
dither
Binary from grayscale indexed
from RGB
Gray2ind
Indexed from grayscale
grayslice
Indexed from grayscale by
thresholding
Im2bw
Binary from intensity, indexed or
RGB image by luminance
threshold.
Ind2gray
Indexed to grayscale
Ind2rgb
Indexed to RGB
Mat2gray
Create a grayscale intensity image
from data in a matrix,
Rgb2gray
RGB to intensity
Rgb2ind
RGB to indexed
Download