Permasalahan Image Processing : Capture, Modelling, Feature Extraction,image Segmentation - Repository UNIKOM
Pertemuan-2:
IMAGE
PROCESSING KK-Komputasi dan Kecerdasan Buatan Teknik Komputer Universitas Komputer Indonesia-UNIKOMJohn Adler Bahan kuliah pertemuan-2
Permasalahan image processing : Capture
modelling, feature extraction,image
segmentation Sejarah Digital Image Processing (DIP)- Beberapa contoh
- Key Stages in Digital Image Processing •
Permasalahan image processing : EXTRACTION, IMAGE SEGMENTATION
Permasalahan capture
Capture merupakan proses awal dari image
- processing untuk mendapatkan gambar Proses capture membutuhkan alat-alat
capture yang baik seperti kamera, scanner,
light-pen dan lainnya, agar diperoleh gambar yang baik.Gambar yang baik akan banyak membantu
- dalam proses selanjutnya.
(inilah sebabnya di semua jurusan fakultas teknik, kuliah matematika menjadi sangat penting !!)
Permasalahan Feature Extraction
Setiap gambar mempunyai karakteristik tersendiri,
- sehingga fitur tidak dapat bersifat general tetapi sangat tergantung pada model dan obyek gambar yang digunakan. Fitur dasar yang bisa diambil adalah warna, bentuk dan
- tekstur. Fitur yang lebih kompleks menggunakan segmentasi, clustering dan motion estimation Pemakaian statistik dan probabilitas, pengolahan sinyal
- sampai pada machine learning diperlukan di sini
Permasalahan Image Segmentation
Bagaimana memisahkan obyek gambar
- dengan backgroundnya
Bagaimana memisahkan setiap obyek
- gambar Teknik clustering apa yang sesuai
- dengan model dan obyek gambar yang digunakan
Contoh Postal Code Problem
History of Digital Image Processing Early 1920s: One of the first applications of
- 00 (2
digital imaging was in the news-
g in ss ce
paper industry
ro P e
- – ag
The Bartlane cable picture
m
I al it
transmission service
ig , D
Early digital image
- – ds oo
Images were transferred by submarine cable
W & z
between London and New York
le za
- – on
Pictures were coded for cable transfer and
G om fr
reconstructed at the receiving end on a telegraph
n ke ta es printer ag Im
History of DIP (cont…) Mid to late 1920s: Improvements to the
- 00 (2
Bartlane system resulted in higher quality
g in ss ce
images
ro P e ag
- – m
New reproduction
I al it
processes based
ig , D ds
on photographic
oo W &
techniques
z le
- – za on
Increased number
G om fr
of tones in
n ke
Improved
ta
reproduced images Early 15 tone digital
es
digital
ag Im
image History of DIP (cont…) 1960s: Improvements in computing technology
- 00 (2
and the onset of the space race led to a surge
g in ss ce
of work in digital image processing
ro P e
- – ag m
1964: Computers used to
I al it
improve the quality of
ig , D ds
images of the moon taken
oo W &
by the Ranger 7 probe
z le
- – za on
Such techniques were used
G om fr
in other space missions
n ke ta
including the Apollo landings
es
A picture of the moon
ag Im
History of DIP (cont…)
1970s: Digital image processing begins to be
- 00 (2
used in medical applications
g in
1979: Sir Godfrey N.
- – ss ce ro P e ag Hounsfield & Prof. Allan M.
m
I al it
Cormack share the Nobel
ig , D ds
Prize in medicine for the
oo W & invention of tomography, z le za
the technology behind
on G om Computerised Axial fr n
Typical head slice CAT
ke
Tomography (CAT) scans
ta es
image
ag Im
History of DIP (cont…)
- 1980s - Today: The use of digital image
processing techniques has exploded and they
are now used for all kinds of tasks in all kinds of
areas- – Image enhancement/restoration
- – Artistic effects
- – Medical visualisation
- – Industrial inspection
- – Law enforcement
- – Human computer interfaces
Beberapa Bidang Ilmu yang Berhubungan
dengan ImageComputer Graphics : Kreasi image
Image processing : penyempurnaan
atau manipulasi gambar- yang hasilnya gambar lainComputer vision : analisis isi image Pengolahan Data Berdasarkan Input/Output
IMAGE Image Processin g
Computer Vision
DESKRIP SI
Computer
Graphics
Data Mining, dllIN P U T
Dua Macam Aplikasi IP Meningkatkan informasi bergambar untuk interpretasi manusia Pemprosesan data image untuk menyimpan, mentransmisikan, dan merepresentasikan untuk persepsi mesin otonomi
Bidang yang Memanfaatkan IP
Berdasarkan sumber dari image: Radiasi dari spektrum elektromagnetik Akustik Ultrasonik Elektronik (dalam bentuk sinar elektron yang digunakan dalam mikroskop elektron) Komputer (image sintetis yang digunakan untuk pemodelan dan visualisasi) Persoalan di dalam IP
- Capture • Modeling • Feature Extraction • Image Segmentation
Permasalahan Capture
Capture (menangkap gambar) merupakan
- proses awal dari image processing untuk mendapatkan gambar Proses capture membutuhkan alat-alat
capture yang baik seperti kamera, scanner,
light-pen dan lainnya, agar diperoleh gambar yang baik.Gambar yang baik akan banyak membantu
- dalam proses selanjutnya.
Hasil Capture : Gamma-Ray Imaging
Nuclear Image :
a. Bone scan
b. PET (Positron Emission Tomography) image.
Astronomical Observations
c. Cygnus Loop Nuclear Reaction d. Radiasi Gamma Hasil Capture : X-Ray Imaging
Hasil Capture : Ultraviolet Imaging
Hasil Capture : Visible Imaging
Hasil Capture : Infrared Imaging
Hasil Capture : Imaging in Microwave Band
Imaging radar : satu- satunya cara untuk menjelajahi daerah yang tidak dapat diakses dari permukaan bumi Radar image dari pegunungan di tenggara Tibet
perhatikan kejelasan dan detail gambar, tidak terhalang oleh awan atau Hasil Capture : Imaging in Microwave Band
Aplikasi ilmu Geologi : eksplorasi mineral dan minyak menggunakan suara dalam spektrum suara rendah (ratusan Hz)
Model seismik dari image cross-sectional
Gambar panah menunjukkan perangkap
(bright spots) hidrokarbon (minyak dan
atau gas)Hasil Capture : Ultrasound Imaging
Peralatan medis :
a. Bayi
b. Melihat bayi dari sisi yang lain c. Thyroids
d. Lapisan tulang menggambarkan lesion Hasil Capture : Imaging in Radio Band
Menggenerate image oleh komputer
Fraktal : an iterative
•
reproduction of basic pattern according to some mathematical rules (a) dan (b) Pemodelan komputer•
3D (c) dan (d)
Examples: Image Enhancement One of the most common uses of DIP
- 00 (2
techniques: improve quality, remove noise etc
g in ss ce ro P e ag m I al it ig , D ds oo W & z le za on G om fr n ke ta es ag Im
Examples: The Hubble Telescope
Launched in 1990 the Hubble
- telescope can take images of very distant objects However, an incorrect mirror
- made many of Hubble’s images useless Image processing
- techniques were used to fix this
Examples: Artistic Effects
- used to make images more visually appealing, to add special effects and to make composite images
Take slice from MRI scan of canine heart, and
- 00 (2
find boundaries between types of tissue
g in
- – ss ce
Image with gray levels representing tissue
ro P e ag
density
m
I
- – al it ig Use a suitable filter to highlight edges , D ds oo W & z le za on G om fr n ke ta es ag Im
- 00
- – (2 g in ss
Digital image processing techniques are used
extensively to manipulate satellite imagery
ce ro P e ag
- – m
Terrain classification
I al
- – it ig Meteorology , D ds oo W & z le za on G om fr n ke ta es ag Im
- 00 (2 g
the World data set
in ss
- – ce ro P Global inventory of e ag m
human settlement
I al it ig
- – , D ds oo
Not hard to imagine
the kind of analysis
W & z le
that might be done
za on G
using this data
om fr n ke ta es ag Im
Examples: Industrial Inspection
- 00 (2 g
expensive, slow and
in ss ce ro P unreliable e ag
I al Make machines do the it ig , D
- m
job instead
ds oo
- W & z le za on
Industrial vision
systems
G om fr
are used in all kinds of
n ke ta es
industries
ag Im
Examples: PCB Inspection
- Machine inspection is used to determine that all
Printed Circuit Board (PCB) inspection
- – components are present and that all solder joints are acceptable Both conventional imaging and x-ray imaging are
- – used
- 00 (2 g
techniques are used
in ss ce ro P extensively by law e ag m
I
enforcers
al it ig
- – , D ds oo
Number plate
recognition for speed
W & z le
cameras/automated
za on G
toll systems
om fr
- – n ke Fingerprint recognition ta es ag
- – Im
Enhancement of CCTV Examples: HCI
- computer interfaces more natural
Face recognition
- – Gesture recognition
- – Does anyone remember the
- user interface from “Minority Report”? These tasks can be
Key Stages in Digital Image Processing
Key Stages in Digital Image Processing
Image Acquisition Image Restoration
Morphological Processing Segmentation
Representation & Description Image Enhancement
Object Recognition Problem Domain Colour Image
Image
(2 g in ss
Key Stages in DIP:Image Aquisition ce ro P e ag m l I
Image Morphological ita
Restoration Processing
ig , D ds ooImage Segmentation
W Enhancement
& z le za
Image Object on
Acquisition Recognition
G om fr n
Representation ke
Problem Domain & Description ta es
Colour Image Image
(2 g in ss
Key Stages in DIP:Image Enhancement ce ro P e ag m l I
Image Morphological ita
Restoration Processing
ig , D ds ooImage Segmentation
W Enhancement
& z le za
Image Object on
Acquisition Recognition
G om fr n
Representation ke
Problem Domain & Description ta es
Colour Image Image
(2 g in ss
Key Stages in DIP: Image Restoration ce ro P e ag m l I
Image Morphological ita
Restoration Processing
ig , D ds ooImage Segmentation
W Enhancement
& z le za
Image Object on
Acquisition Recognition
G om fr n
Representation ke
Problem Domain & Description ta es
Colour Image Image
(2 g in
Key Stages in Digital Image Processing: ss ce
Morphological Processing ro P e ag m l I
Image Morphological ita
Restoration Processing
ig , D ds ooImage Segmentation
W Enhancement
& z le za
Image Object on
Acquisition Recognition
G om fr n
Representation ke
Problem Domain & Description ta es
Colour Image Image
(2 g in ss
Key Stages in DIP: Segmentation ce ro P e ag m l I
Image Morphological ita
Restoration Processing
ig , D ds ooImage Segmentation
W Enhancement
& z le za
Image Object on
Acquisition Recognition
G om fr n
Representation ke
Problem Domain & Description ta es
Colour Image Image Key Stages in DIP: Object Recognition
(2 g
Restoration Processing
in ss ce ro P e ag m
I Image al
Segmentation
it
Enhancement
ig , D ds oo W &
Image Object
z le
Acquisition Recognition
za on G om fr n
Representation
ke
Problem Domain
ta
& Description
es ag Im
Colour Image Image
(2 g in
Key Stages in Digital Image Processing: ss ce
Representation & Description ro P e ag m l I
Image Morphological ita
Restoration Processing
ig , D ds ooImage Segmentation
W Enhancement
& z le za
Image Object on
Acquisition Recognition
G om fr n
Representation ke
Problem Domain & Description ta es
Colour Image Image Key Stages in Digital Image Processing:
Image Compression
Image Morphological
Restoration Processing
ImageSegmentation Enhancement Image
Object Acquisition Recognition Representation
Problem Domain & Description Colour Image Image Key Stages in Digital Image Processing:
Colour Image Processing
Image Morphological
Restoration Processing
ImageSegmentation Enhancement Image
Object Acquisition Recognition Representation
Problem Domain & Description Colour Image Image
Penutup
Ada beberapa hal yang harus dikuasai sebelum menguasai
materi di dalam image processing
yaitu : matematika, aljabar, pengolahan sinyal, matriks dan transformasi linier, statistik,Referensi slide
Brian Mac Namee, Digital Image Processing :
- Introductio Nana Ramadijanti, Image Processing : Day-1,
- Laboratorium Computer Vision, PENS-ITS, Surabaya Achmad Basuki, Pengantar Pengolahan •
Citra, Laboratorium Computer Vision, PENS-
ITS, Surabaya
Image Processing Mempelajari Apa?
- http://www.mathworks.com/products/
- image/description4.html
- http://www.mathworks.com/products/image/
- description7.html
INTRODUCTION In electrical engineering and computer science image
processing is any form of signal processing for which the input
is an image, such as a photograph or video frame; the output of image processing may be either an image or, a set ofcharacteristics or parameters related to the image. Most image-
processing techniques involve treating the image as a two- dimensional signal and applying standard signal-processing techniques to it.
Image processing usually refers to digital image processing, but
optical and analog image processing also are possible. This article is about general techniques that apply to all of them. The acquisition of images (producing the input image in the first place) is referred to as imaging.Definitions of Image Processing •The general term "image processing" refers to a computer discipline wherein digital images are the main data object. This type of processing can be broken down into several sub-categories, including: compression, image enhancement,
image filtering, image distortion, image display
and coloring.•Any activity that transforms an input image into
an output image. Manipulation of an image to improve or change some quality of the imageThis Process involves two aspects
Improving the visual appearance of images to
a human viewer.Preparing images for measurement of the features and structures present.
Why do we need it……?
Since the digital image is invisible it must
- prepare for viewing one or more output device. The digital image can be optimized for the application by enhancing or altering the appearance of the structures within it.
It might be possible to analyze the image in
- the computer and provide clues to the radiologist to help direct important/suspicious structure.
Acquiring Images
Since the digital image is
- invisible it must prepare for viewing one or more output
device. The digital image can
be optimized for the application by enhancing or altering theappearance of the structures
within it. It might be possible to analyze - the image in the computer and provide clues to the radiologist to help direct important/
High Resolution
The process of obtaining a high resolution (HR) image or a
sequence of HR from a set of low resolution (LR) observation. HR technique has applied to a variety of fields such as obtaining.Improve still images. High definition television.
High performance color liquid crystal display (LCD) screen.
Video surveillance. Remote sensing and Medical imaging.Color Spaces
Conversion from RGB (The brightness of
- individual red, green and green signal at defined wavelength) to YIQ/YUV and to other color encoding schemes is straightforward and losses no information.
Image Sensors
Digital processing requires images to be
- obtained in the form of electrical signals. These signals can be digitized into sequence of numbers which can be processed by a computer.
Image Intensity Equalization using Histograms
Image intensity Equalization is the process of converting the given image into the
desired manner using Histogram. In histogram equalization we are trying to maximize the image contrast by applying a gray level transform which tries to flatten the resulting histogram. The gray level transform is a scaled version of the original image's cumulative histogram. That is, the gray level transform T is given by T[i] = (G-1)c(i), where G is the number of gray levels and c(i) is the normalized cumulative histogram of the original image.When we want to specify a non-flat resulting histogram, we can use the following steps: Specify the desired histogram g(z) Obtain the transform which would equalize the specified histogram, Tg, and its inverse Tg
- 1
Get the transform which would histogram equalize the original image, s=T[i] Apply the inverse transform Tg
- 1
- 1
on the equalized image, that is z=Tg
[s] Input Image corresponding histogram Output Image corresponding histogram
Multiple Images: It may constitute a series of views of the same area using different wavelength of light and other signals. Examples
includes the image processed by satellites those images may
require processing. Hardware Requirement: A general purpose computer can be used for image processing; four key demands must be met- high resolution image display • Sufficient memory transfer bandwidth.
- Sufficient storage space
Adobe The Image processing is used to get/acquire
image in a desired manner by using some of the software’s
shown above without affecting its original input image and
their may not be loos of data during the conversion process.