18
IF-UTAMA 4/4/2012 1 Format File Bitmap Dosen Pembina : Sriyani Violina, M.T. Danang Junaedi 1 RASTER GRAPHICS 2 alfeacamia.cmswiki.wikispaces.net/file/view/2.01+Raster+Graphics.ppt Raster Graphics Also called bitmap graphics Consist of grids of tiny dots called pixels Have a fixed resolution and cannot be resized without altering image quality Edited in paint programs Bitmap enlargement Notice the pixels Image source: http://graphicssoft.about.com/od/aboutgraphics/a/bitmapvector.htm 3 Common Raster Formats • GIF • JPEG • BMP • PNG • TIFF 4

Format File Bitmap - · PDF fileFormat File Bitmap Dosen Pembina : ... • Microsoft has defined a particular representation of color bitmaps of ... menyediakan fungsi

Embed Size (px)

Citation preview

IF-UTAMA 4/4/2012

1

Format File Bitmap

Dosen Pembina :

Sriyani Violina, M.T.

Danang Junaedi

1

RASTER GRAPHICS

2

alfeacamia.cmswiki.wikispaces.net/file/view/2.01+Raster+Graphics.ppt

Raster Graphics

• Also called bitmap

graphics

• Consist of grids of tiny dots called pixels

• Have a fixed resolution and cannot

be resized without altering image quality

• Edited in paint

programs Bitmap enlargement

Notice the pixels

Image source: http://graphicssoft.about.com/od/aboutgraphics/a/bitmapvector.htm

3

Common Raster Formats

• GIF

• JPEG

• BMP

• PNG

• TIFF

4

IF-UTAMA 4/4/2012

2

GIF – Graphics Interchange Format

� Animation – Standard

format for animation on

the Internet.

� Transparency – yes

• Lossless compression

• Colors = 256 (8-bit)

• Most common format for:

– Text

– Clip art, animations, icons, logos

– Simple diagrams, line drawings

– Graphics with large blocks

of a single color

– Graphics with transparent

areas

– Images displayed on computer screens and on

websites.Animated Gif

5

JPEG – Joint Photographic Experts Group

X Animation – No

X Transparency – No

• Lossy compression

• Colors – 16.7 M (24-bit)

• High quality but larger file

size than a GIF

• Commonly Used For:

– Desktop publishing photographs

– Photographs and natural

artwork

– Scanned photographs

– Emailing photographs

– Digital camera photographs

6

BMP - Bitmap

X Animation – No

X Transparency – No

• Uncompressed

• 256 colors

• Large file size - not well suited for transfer

across the Internet or for print publications

• Commonly Used For:

– Editing raster graphics

– Creating icons and

wallpaper

– On-screen display

Icons 7

PNG – Portable Network Graphics

X Animation – no

� Transparency – yes

• Lossless compression

• 256 colors

– Not suited for

photographs

• Commonly Used For:

– Replacing GIF and

TIFF images

– Online viewing of

images

• See examples at

http://graphicssoft.about.com/od/freedownlo

ads/l/blfreepng07.htm

8

IF-UTAMA 4/4/2012

3

TIFF – Tagged Image File Format

X Animation – No

X Transparency – No

• Available in compressed

and un-compressed

formats

• Compressed is advised

• Colors – 16 M (24-bit)

• Commonly Used For:

– Storage container for faxes and other digital images

– To store raw bitmap data

by some programs and devices such as scanners

– High resolution printing

– Desktop Publishing images

9

Demonstration of File Sizes

10samsclass.info/131/pptF05/ch09.ppt

BITMAP

11

1. pages.towson.edu/hilberg/COSC109/Ch06.ppt

2. Referensi lain yang terkait

BMP File Bitmap

• The BMP file format, sometimes called bitmap or DIB file format (for

device-independent bitmap), is an image file format used to store bitmap

digital images, especially on Microsoft Windows and OS/2 operating

systems.

• Bitmap is derived from the words ‘bit’, which means the simplest element in

which only two digits are used, and ‘map’, which is a two-dimensional matrix

of these bits. A bitmap is a data matrix describing the individual dots or

pixels (picture elements) of an image.

• Bitmapped images are known as paint graphics.

• Bitmapped images can have varying bit and color depths.

• Bitmaps are an image format suited for creation of:

– Photo-realistic images.

– Complex drawings.

– Images that require fine detail.

• Many graphical user interfaces use bitmaps in their built-in graphics subsystems;[1] for example, the Microsoft Windows and OS/2 platforms' GDI subsystem, where the specific format used is the Windows and OS/2 bitmap file format, usually named with the file extension of .BMP or .DIB.

12

IF-UTAMA 4/4/2012

4

Bitmaps

13

Available binary Combinations for Describing a

Color

Bitmaps

1. 24 bit color (millions of colors)

2. 8 bit color (256 colors)

3. 8 bit color (Mac optimized palette)

4. 4 bits color (16 colors)

5. 8 bit gray scale (256 shades)

6. 4 bit gray scale (16 shades)

7. 1 bit gray scale (2 shades)

14

Bitmaps

Bitmaps can be inserted by:

– Using clip art galleries - an assortment of graphics, photographs, sound, and video. A popular alternative for users who do not want to create their own images.

– Using bitmap software such as Adobe's Photoshop and Illustrator, Macromedia's Fireworks, Corel's Painter, CorelDraw, Quark Express.

– Capturing and editing images.

• Capturing and storing images directly from the screen is another

way to assemble images for multimedia.

• Image editing enables one to enhance and make composite images,

alter and distort images and add and delete elements.

– Scanning images from conventional sources and make necessary alterations and manipulations.

15

Device-independent bitmaps and BMP file

format

BMP File Header Stores general information about the BMP file.

Bitmap Information(DIB header)

Stores detailed information about the bitmap image.

Color PaletteStores the definition of the colors being used for indexed color bitmaps.

Bitmap Data Stores the actual image, pixel by pixel.

16

• Microsoft has defined a particular representation of color bitmaps of different color depths, as an aid to exchanging bitmaps between

devices and applications with a variety of internal representations. They called these device-independent bitmaps or DIBs, and the file format for them is called DIB file format or BMP file format.

• A typical BMP file usually contains the following blocks of data:

IF-UTAMA 4/4/2012

5

BMP file header• This block of bytes is at the start of the file and is used to identify the file.

• A typical application reads this block first to ensure that the file is actually a BMP file and that it is not damaged.

• Note that the first two bytes of the BMP file format (thus the BMP header) are stored in big-endian order.

• This is the magic number 'BM'. All of the other integer values are stored in little-endian format (i.e. least-significant byte first)

Offset# Size Purpose

0000h 2 bytes

the magic number used to identify the BMP file: 0x42 0x4D (Hex code points for B and M). The following entries are possible: •BM - Windows 3.1x, 95, NT, ... etc

•BA - OS/2 Bitmap Array •CI - OS/2 Color Icon •CP - OS/2 Color Pointer •IC - OS/2 Icon •PT - OS/2 Pointer

0002h 4 bytes the size of the BMP file in bytes

0006h 2 bytes reserved; actual value depends on the application that creates the image

0008h 2 bytes reserved; actual value depends on the application that creates the image

000Ah 4 bytesthe offset, i.e. starting address, of the byte where the bitmap data can be found.

17

Bitmap information (DIB header)

This block of bytes tells the application detailed information about the image, which will be used to display the image on the screen. The

block also matches the header used internally by Windows and OS/2 and has several different variants.

Size Header Identified by Supported by the GDI of

12 OS/2 V1BITMAPCOREH

EADER

OS/2 and also all Windows

versions since Windows 3.0

64 OS/2 V2BITMAPCOREH

EADER2

40 Windows V3BITMAPINFOHE

ADER

all Windows versions since

Windows 3.0

108 Windows V4BITMAPV4HEAD

ER

all Windows versions since

Windows 95/NT4

124 Windows V5BITMAPV5HEAD

ERWindows 98/2000 and newer

18

BITMAPFILEHEADER

Offset Size Name Description

0 2 bfType ASCII “BM”

2 4 bfSize Size of file (in bytes)

6 2 bfReserved1 Zero

8 2 bfReserved2 Zero

10 4 bfOffBits Byte offset in files where image begins

14 4 biSize Size of this header (40 bytes)

18 4 biWidth Image width in pixels

22 4 biHeight Image height in pixels

26 2 biPlanes Number of image planes, must

be 1

28 2 biBitCount Bits per pixel: 1,4,8, or 24

30 4 biCompression Compression type

19

BITMAPINFOHEADER (Windows 3) (cont’d)

Offset Size Name Description

34 4 biSizeImage Size of compressed image (in bytes),

zero if uncompressed

38 4 biXPelsPerMeter Horizontal resolution (pixels/meter)

42 4 biYPelsPerMeter Vertical resolution (pixels/meter)

46 4 biClrUsed Number of colors used

50 4 biClrImportant Number of ‘important’ color

54 4*N bmiColors Color map

20

IF-UTAMA 4/4/2012

6

Proses Pembacaan Citra 24 bit

21

Windows RGBQUAD

Offset Name Description

0 rgbBlue Blue value for color map entry

1 rgbGreen Green value

2 rgbRed Red value

3 rgbReserved Zero

22

Proses Penentuan Warna Ke Layar

• Untuk file 24 bit Informasi intensitas RGB sudah dapat langsung diketahui dari bitmap data, sedangkan untuk file 1,4,8 bit informasi

RGB diperoleh dari Color Map

23

Proses Penentuan Warna Ke Layar

• Pada umumnya setiap bahasa pemrograman telah

menyediakan fungsi untuk menghasilkan warna apabila

kita telah mengetahui intensitas RGB:

– Contoh dalam delphi:

• Image1.canvas.pixel(1,1)=RGB(10,8,2);

– Contoh dalam Visual Basic:

• PicBaru.PSet (SbX, SbY), RGB(10, 8, 2)

24

IF-UTAMA 4/4/2012

7

A

B

• Perlu diperhatikan bahwa dalam file data disimpan dari

belakang ke depan secara sequential. Berarti bitmap

data pertama adalah pixel pada posisi A dan bitmap data

terakhir adalah pixel pada posisi B

Penentuan Posisi Pixel

25

Color map

• Citra 1, 4, dan 8 bit per pixel butuh color map

• Entri dalam color map (palette) biasanya 2, 16, atau 256

– Bisa lebih sedikit jika citra tidak membutuhkan semua warna yang tersedia

– Jumlah warna yang digunakan = biClrUsed

– biClrUsed = 0 � color map memuat semua warna

– 4 byte per entri

• Entri awal color map = warna penting

– Jumlah warna penting = biClrImportant � jumlah warna yang diperlukan untuk mendapat tampilan citra yang cukup bagus

26

Proses Pembacaan Citra 8 bit

• Citra dengan kedalaman 8 bit berarti 1 pixel diwakili oleh 1 byte dan memiliki kemungkinan warna sebanyak 8 bit

• Prosesnya sama dengan pembacaan citra 24 bit dimana kita membaca :

• FileHeader sebesar 14 byte

• InfoHeader 40 byte

• ColorMap

• Bitmap Data

27

Proses Pembacaan Citra 8 bit

Dengan mengetahui informasi

mengenai OffBits maka kita bisa

menghitung posisi offset dari

ColorMap yaitu dimulai dari offset

54 sampai dengan nilai yang

tersimpan didalam offbits(X)

28

IF-UTAMA 4/4/2012

8

Proses Pembacaan Citra 8 bit

• Analogi Color Map adalah mengindex warna yang ada

ke dalam tabel sehingga bitmap data tidak lagi berisi

data intensitas RGB namun mengandung index warna

• Untuk mengetahui warna pixel(x) maka kita mengakses

color map dengan index sesuai dengan nilai yang

tersimpan pada bitmap data

29

Proses Pengambilan Warna dari Color Map

Berarti untuk pixel 1 intensitas

RGB : 56 5 9

Berarti untuk pixel 2 intensitas

RGB : 5 34 67

Berarti untuk pixel 3 intensitas

RGB : 5 34 67

COLORMAP

B 0RG B 0RG

BGR

0 9556

1 67345

15 67345

30

Menentukan Ukuran File dari Bitmap

• Yang membedakan antara citra 1,4,8,24 bit adalah ukuran storage yang digunakan untuk menyimpan warna dari 1 buah pixel

• Misalkan: citra A :200 x 200 pixel

– Hitung berapa minimum byte dari file bitmap yang dihasilkan bila:

a. citra A disimpan dalam 8 bit

b. citra A disimpan dalam 24 bit

– Solusi

a. 200 x 200 x 1 + 54 + 256 * 3 = 40822 byte

b. 200 x 200 x 3 + 54 = 120054 byte

31

EQUALISASI HISTOGRAM

SPESIFIKASI HISTOGRAM

32

IF-UTAMA 4/4/2012

9

Dua Pendekatan Image Enhancement

• Metode-metode berbasis domain frekwensi

– Manipulasi terhadap representasi frekwensi dari citra

– Contoh: operasi berbasis transformasi Fourier terhadap citra

• Metode-metode berbasis domain spasial

– Manipulasi langsung terhadap pixel-pixel pada citra

– Contoh: operasi histogram

33

Histogram citra

• Berlaku untuk nilai gray level; RGB � per plane warna

• Plotting dari persamaan:

– L: jumlah level

– pr(rk): probabilitas kemunculan level ke-k

– nk: jumlah kemunculan level k pada citra

– n: total jumlah pixel dalam citra

1,...,1,0;10;)( −=≤≤= Lkrn

nrp k

kkr

34

Contoh histogram

35

Equalisasi histogram

• Tujuan: melakukan transformasi terhadap histogram citra

asli sedemikian sehingga didapat histogram citra hasil

dengan distribusi lebih seragam (uniform) ≈ linearisasi

• Dasar konsep: transformasi probability density function

menjadi uniform density � bentuk kontinyu

• Agar dapat dimanfaatkan dalam pengolahan citra digital,

diubah ke bentuk diskrit

36

IF-UTAMA 4/4/2012

10

Equalisasi pada domain kontinyu

[ ] 1011)(

1)()(

:

10;)()(:

)()(:

)(

)(

0

)(

1

1

1

≤≤==

=

≤≤==

=

=

=

=

srp

rpsp

Uniform

rdwwpsTssiTransforma

ds

drrpspHistogram

sTr

sTrr

rs

r

r

sTr

rs

37

Ilustrasi equalisasi pada domain kontinyu

38

Bentuk diskrit fungsi transformasi

10)(

1,...,1,0

10)()(

1

0 0

≤≤=

−=

≤≤===

= =

∑ ∑

kkk

kk

j

k

j

jr

j

kk

ssTr

Lk

rrp

n

nrTs

39

Operasi equalisasi histogram

1. Buat histogram dari citra asli

2. Transformasikan histogram citra asli menjadi

histogram dengan distribusi seragam

3. Ubah nilai tiap pixel sesuai dengan nilai hasil pemetaan (histogram asli � uniform histogram)

40

IF-UTAMA 4/4/2012

11

Pseudo Code : citra 512 x 512 pixel 256 graylevel

Var x,y,i,j : integer;

HistEq : array[0..255] of integer;

Hist : array[0..255] of real;Sum : real;

Begin

Histogram(image,Hist) {bentuk histogram dari citra asli}

for i:= ∅∅∅∅ to 255 do {transformasi ke uniform histogram}

sum := 0.0

for j:= ∅∅∅∅ to i dosum:= sum + hist[j]

endfor

histEq[i]:=round(255 * sum);

end;for y:=0 to 511 do {ubah nilai tiap pixel pada citra}

for x:=0 to 511 do

image[x,y]:= HistEq[Image[x,y]];end;

end;

end;

41

Contoh

Citra 64x64 pixel, 8 tingkat keabuan dgn distribusi:

rk nk pr(rk)=nk/n

r0=0 790 0,19

r1=1/7 1023 0,25

r2=2/7 850 0,21

r3=3/7 656 0,16

r4=4/7 329 0,08

r5=5/7 245 0,06

r6=6/7 122 0,03

r7=1 81 0,02

Histogram citra:

0

0,05

0,1

0,15

0,2

0,25

0,3

0 1/7 2/7 3/7 4/7 5/7 6/7 1

gray level (rk)

pro

ba

bil

ity (

pr(

r k))

42

Fungsi transformasi

00.1)()(

98.0)()(;95.0)()(

89.0)()(;81.0)()(

65.0)()()()()(

44.0)()()()(

19.0)()()(

7

0

77

6

0

66

5

0

55

4

0

44

3

0

33

210

2

0

22

10

1

0

11

0

0

0

00

===

======

======

=++===

=+===

====

∑∑

∑∑

=

==

==

=

=

=

j

jr

j

jr

j

jr

j

jr

j

jr

rrr

j

jr

rr

j

jr

r

j

jr

rprTs

rprTsrprTs

rprTsrprTs

rprprprprTs

rprprprTs

rprprTs

43

Fungsi transformasi: grafik

0

0,2

0,4

0,6

0,8

1

1,2

0 1/7 2/7 3/7 4/7 5/7 6/7 1

gray level (rk)

tran

sfo

rmed

valu

e (

sk)

44

IF-UTAMA 4/4/2012

12

Pembulatan

• 8 tingkat keabuan valid � nilai Sk dibulatkan ke nilai

valid terdekat [Normal (Sk x (tingkat kedalaman-1))]

� s0 = 0.19 ≅ 1/7

� s1 = 0.44 ≅ 3/7

� s2 = 0.65 ≅ 5/7

� s3 = 0.81 ≅ 6/7

� s4 = 0.89 ≅ 6/7

� s5 = 0.95 ≅ 1

� s6 = 0.98 ≅ 1

� s7 = 1.00 ≅ 1

45

Pemetaan

• Hanya ada 5 level keabuan pada uniform histogram

– r0 (790 pixel) � s0 = 1/7

– r1 (1023 pixel) � s1 = 3/7

– r2 (850 pixel) � s2 = 5/7

– r3 (656 pixel), r4 (329 pixel) � s3 = 6/7

– r5 (245 pixel),r6 (122 pixel),r7 (81 pixel) � s4 = 7/7

46

Histogram dengan distribusi seragam

0

0,05

0,1

0,15

0,2

0,25

0,3

0 1/7 2/7 3/7 4/7 5/7 6/7 1

gray leve l (sk)

pro

bab

ilit

y (

ps(s

k))

�Karena histogram merupakan aproksimasi terhadap

probability density function, sangat jarang didapat

histogram hasil yang betul-betul rata

47

Tabel Histogram secara Lengkap

rk nk pr(rk)=nk/n Sk Sk x 7 Normal(Sk)

r0=0 790 0,19 0,19 1,33 ≅≅≅≅ 1 s0=1/7

r1=1/7 1023 0,25 0,44 3,08 ≅≅≅≅ 3 s1=3/7

r2=2/7 850 0,21 0,65 4,55 ≅≅≅≅ 5 s2=5/7

r3=3/7 656 0,16 0,81 5,67 ≅≅≅≅ 6 s3=6/7

r4=4/7 329 0,08 0,89 6,23 ≅≅≅≅ 6 s4=6/7

r5=5/7 245 0,06 0,95 6,65 ≅≅≅≅ 7 s5=7/7

r6=6/7 122 0,03 0,98 6,86 ≅≅≅≅ 7 s6=7/7

r7=1 81 0,02 1,00 7 s7=1

48

Citra 64x64 pixel, 8 tingkat keabuan dgn distribusi:

IF-UTAMA 4/4/2012

13

Hasil Equalisasi

rj�sk nk ps(sk)

r0�s0=1/7 790 0,19

r1�s1=3/7 1023 0,25

r2�s2=5/7 850 0,21

r3,r4 �s3=6/7 985 0,24

r5,r6,r7 �s4=7/7 448 0,11

0

0,05

0 ,1

0 ,15

0 ,2

0 ,25

0 ,3

0 1/7 2 /7 3 /7 4 /7 5 /7 6 /7 1

g ra y le ve l (sk )

pro

bab

ilit

y (

ps(s

k))

49

Contoh1 equalisasi histogram

50

Contoh 2 equalisasi histogram

51

Spesifikasi histogram

• Kelemahan equalisasi histogram:

histogram hasil tidak bisa dibentuk sesuai kebutuhan

• Kadangkala dibutuhkan untuk lebih

menonjolkan rentang gray level tertentu pada citra � spesifikasi histogram

52

IF-UTAMA 4/4/2012

14

Operasi spesifikasi histogram

1. Buat histogram dari citra asli

2. Transformasikan histogram citra asli menjadi histogram dengan distribusi seragam

3. Tentukan fungsi trasformasi sesuai spesifikasi histogram yang diinginkan

4. Ubah nilai tiap pixel sesuai dengan nilai hasil pemetaan (histogram asli � histogram equalisasi � histogram hasil)

53

Algoritma: citra 512 x 512 pixel 256 graylevel

Var x,y,i,minval,minj,j : integer; Histspec : array[0..255] of integer;Invhist : array[0..255] of integer; Sum : real;

BeginHist_Equalization(Image) {equalisasi histogram}

For i:= 0 to 255 do {histogram yang dispesifikasikan telah disimpan di spec}

Sum:= 0.0;For j:= 0 to i do Sum := sum + spec[j]

Histspec[i] = round(255 * sum)Endfor {didapat fungsi transformasi}

for i:= 0 to 255 do {pemetaan histogram}

minval := abs(i – histspec[0]; minj := 0;

for j:= 0 to 255 doif abs(i – histspec[j]) < minval then

minval := abs(i – histspec[j])minj := j

endifinvhist[i]:= minj

endforendfor

for y:= 0 to 511 do {ubah nilai tiap pixel pada citra}

for x:= 0 to 511 do image[x,y] = invhist[image(x,y)]

54

Bentuk diskrit spesifikasi histogram: by example

Citra 64x64 pixel, 8 tingkat keabuan dgn distribusi:

rk nk pr(rk)=nk/n

r0=0 790 0,19

r1=1/7 1023 0,25

r2=2/7 850 0,21

r3=3/7 656 0,16

r4=4/7 329 0,08

r5=5/7 245 0,06

r6=6/7 122 0,03

r7=1 81 0,02

Histogram citra:

0

0,05

0,1

0,15

0,2

0,25

0,3

0 1/7 2/7 3/7 4/7 5/7 6/7 1

gray level (rk)

pro

ba

bil

ity (

pr(

r k))

55

Bentuk histogram yang diinginkan

zk pz(zk)

z0=0 0,00

z1=1/7 0,00

z2=2/7 0,00

z3=3/7 0,15

z4=4/7 0,20

z5=5/7 0,30

z6=6/7 0,20

z7=1 0,15

0,00

0,05

0,10

0,15

0,20

0,25

0,30

0,35

0 1/7 2/7 3/7 4/7 5/7 6/7 1

gray level (zk)

pro

bab

ilit

y (

pz(z

k))

56

IF-UTAMA 4/4/2012

15

Langkah 1: equalisasi histogramDidapat hasil:

rj�sk nk ps(sk)

r0�s0=1/7 790 0,19

r1�s1=3/7 1023 0,25

r2�s2=5/7 850 0,21

r3,r4 �s3=6/7 985 0,24

r5,r6,r7 �s4=7/7 448 0,11

0

0,05

0 ,1

0 ,15

0 ,2

0 ,25

0 ,3

0 1/7 2 /7 3 /7 4 /7 5 /7 6 /7 1

g ra y le ve l (sk )

pro

bab

ilit

y (

ps(s

k))

57

Langkah 2: cari fungsi transformasi

∑=

==k

j

jzkk zpzGv0

)()(

� v0 = G(z0) = 0,00

� v1 = G(z1) = 0,00

� v2 = G(z2) = 0,00

� v3 = G(z3) = 0,15

� v4 = G(z4) = 0,35

� v5 = G(z5) = 0,65

� v6 = G(z6) = 0,85

� v7 = G(z7) = 1,00

58

Langkah 1 : Equalisasi

rk nk pr(rk)=nk/n Sk Sk x 7 Normal(Sk)

r0=0 790 0,19 0,19 1,33 ≅ 1 s0=1/7

r1=1/7 1023 0,25 0,44 3,08 ≅ 3 s1=3/7

r2=2/7 850 0,21 0,65 4,55 ≅ 5 s2=5/7

r3=3/7 656 0,16 0,81 5,67 ≅ 6 s3=6/7

r4=4/7 329 0,08 0,89 6,23 ≅ 6 s4=6/7

r5=5/7 245 0,06 0,95 6,65 ≅ 7 s5=7/7

r6=6/7 122 0,03 0,98 6,86 ≅ 7 s6=7/7

r7=1 81 0,02 1,00 7 s7=1

59

Langkah 2: cari fungsi transformasi

zk pz(zk) Vk Vk x 7 Normal(Vk)

z0=0 0,00 0,00 0,00 v0=0

z1=1/7 0,00 0,00 0,00 v1=0

z2=2/7 0,00 0,00 0,00 v2=0

z3=3/7 0,15 0,15 1,05 ≅ 1 v3=1/7

z4=4/7 0,20 0,35 2,45 ≅ 2 v4=2/7

z5=5/7 0,30 0,65 4,45 ≅ 4 v5=4/7

z6=6/7 0,20 0,85 5.95 ≅ 6 v6=6/7

z7=1 0,15 1,00 7 v7=1

Dengan kata lain, lakukan langkah-langkah equalisasi

thd histogram yang diinginkan :

60

IF-UTAMA 4/4/2012

16

Grafik fungsi transformasi

0

0,2

0,4

0,6

0,8

1

1,2

0 1/7 2/7 3/7 4/7 5/7 6/7 1

gray level (zk)

tran

sfo

rmati

on

(v

k)

61

Langkah 3: terapkan inverse G pada level histogram equalisasi

Pemetaan nilai sk ke G(zk) terdekat

� s0 = 1/7 ≈ 0.14 � G(z3) = 0.15; z3 = 3/7

� s1 = 3/7 ≈ 0.43 � G(z4) = 0.35; z4 = 4/7

� s2 = 5/7 ≈ 0.71 � G(z5) = 0.65; z5 = 5/7

� s3 = 6/7 ≈ 0.86 � G(z6) = 0.85; z6 = 6/7

� s4 = 1 � G(z7) = 1.00; z7 = 1

62

Langkah 4: pemetaan dari rk ke zk

• Dengan memperhatikan pemetaan histogram asli ke

histogram equalisasi

� r0 = 0 � z3 = 3/7

� r1 = 1/7 � z4 = 4/7

� r2 = 2/7 � z5 = 5/7

� r3 = 3/7 � z6 = 6/7

� r4 = 4/7 � z6 = 6/7

� r5 = 5/7 � z7 = 1

� r6 = 6/7 � z7 = 1

� r7 = 1 � z7 = 1

63

Histogram hasil

zk nk pz(zk)=nk/n

r0=0 0 0

r1=1/7 0 0

r2=2/7 0 0

r3=3/7 790 0,19

r4=4/7 1023 0,25

r5=5/7 850 0,21

r6=6/7 985 0,24

r7=1 448 0,11

0,00

0,05

0,10

0,15

0,20

0,25

0,30

0 1/7 2/7 3/7 4/7 5/7 6/7 1

gray level (zk)

pro

bab

ilit

y (

pz(z

k))

Histogram hasil mungkin tidak sama persis dengan spesifikasinya � transformasi

hanya akan memberikan hasil yang persis pada kasus kontinyu

64

IF-UTAMA 4/4/2012

17

Contoh 1 spesifikasi histogram

65

Contoh 2 spesifikasi histogram

66

Contoh 3 spesifikasi histogram

67

Contoh cara menspesifikasikan histogram

68

IF-UTAMA 4/4/2012

18

Local enhancement

• Metode equalisasi dan spesifikasi

histogram yg telah dibahas bersifat global (operasi terhadap semua pixel dalam citra)

• Kadang diperlukan enhancement hanya

untuk suatu area tertentu dalam citra

– Adaptasi metode global (equalisasi atau spesifikasi) untuk area N x M pixel

69