GNR401 Dr. A.

Bhattacharya 1

FUNDAMENTALS OF
DIGITAL IMAGES
Lecture 2
 Image Data Structures
2

Common Data Structures to Store

Multiband Data

 BIL – band interleaved by line

 BSQ – band sequential
 BIP – band interleaved by pixel

GNR401 Dr. A. Bhattacharya

 Example
3

1 1 1 2 2 2 3 3 3

1 1 1 2 2 2 3 3 3

1 1 1 2 2 2 3 3 3

Band 1 Band 2 Band 3

3 bands , 9 pixels each in (3x3 format)

GNR401 Dr. A. Bhattacharya

 BIL
4

 Band interleaved by line storage format

 MxN Image; K Bands; One row on ground
B11 B12 … B1N
B21 B22 … B2N
…
Bk1 Bk2 … BkN

 A single file on disk or CD contains M.K rows, each having

N columns; Every K rows in the file correspond to ONE ROW
ON THE GROUND

GNR401 Dr. A. Bhattacharya

 BIL
5

Band 1 Row1
… Image Size
Band K Row1 M rows
BIL FILE
STRUCTURE Band1 Row2 N columns
… K Bands
Band K Row2
…
Band 1 Row M
…
Band K Row M
GNR401 Dr. A. Bhattacharya
 1
1 1 1 1
1
6 1 1 1 2
2
2
1 1 1 3  Line #1, band #1 is
3
3 stored first
1
2 2 2 1
1
 Followed by line #1,
2 2 2
2
2
band #2
2
2 2 2 3  Bands are inter-leaved
3
3 by line
1
3 1
3 3 1

3 3 3
2
2
 BIL format
2
3
3 3 3 3
3
GNR401 Dr. A. Bhattacharya
 BIL
7

 BIL is a popular format for storing multispectral

images, and supported by most remote sensing
software (ERDAS, PCI, …)
 Well suited when multiband data analysis is
required
 Lot of data I/O involved when access to a single
band image is needed on sequential access
systems. Moderate overhead on random access
systems

GNR401 Dr. A. Bhattacharya

 BSQ
8

 Band sequential method involves storing one full

single band image after another
B11 B12 … B1N
B21 B22 … B2N
…
BM1 BM2 … BMN

 The image for the second band, …, up to Band K

follow

GNR401 Dr. A. Bhattacharya

 BSQ
9

Band 1 Row 1
…
Band 1
Band 1 Row M
Band2 Row 1
Image Size … Band 2
Band 2 Row M
M rows …
N columns Band K Row 1
K Bands … Band K
Band K Row M

GNR401 Dr. A. Bhattacharya

 1
1 1 1 1
1
1
10 1 1 1
1
1
1 1 1 1  Band #1 is stored first
1
1
2
 Followed by #2, #3
2
2 2 2  Bands are stored
2
2
2 2 2 2 sequentially
2
2
2 2 2 2
2
3  Band sequential (BSQ)
3
3 3 3 3 format
3
3 3 3 3
3
3
3 3 3 3
3
GNR401 Dr. A. Bhattacharya
 BSQ
11

 Ideally suited when the multiband image is

processed one band at a time, such as image
enhancement, neighbourhood filtering, etc.
 More overheads when all band values are required
at each pixel

GNR401 Dr. A. Bhattacharya

 BIP
12

 Band interleaved by pixel

 Commonly used for storing color images, with red, green
and blue values alternating
R GBRGBRGB…
 Notused in present times to store satellite images
 Used in the early stages of Landsat data distribution

GNR401 Dr. A. Bhattacharya

 BIP
13

First Row
Band 1 Band 2 … Band K Band 1 Band 2 … Band K … Band K
Row 1 Row 1 Row 1 Row 1 Row 1 Row 1 Row 1
Pixel 1 Pixel 1 … Pixel 1 Pixel 2 Pixel 2 Pixel 2 Pixel N

Second Row
Band 1 Band 2 … Band K Band 1 Band 2 … Band K … Band K
Row 2 Row 2 Row 2 Row 2 Row 2 Row 2 Row 2
Pixel 1 Pixel 1 … Pixel 1 Pixel 2 Pixel 2 Pixel 2 Pixel N
…
Mth Row
Band 1 Band 2 … Band K Band 1 Band 2 … Band K … Band K
Row M Row M Row M Row M Row M Row M Row M
Pixel 1 Pixel 1 … Pixel 1 Pixel 2 Pixel 2 Pixel 2 Pixel N
GNR401 Dr. A. Bhattacharya
 1
1 1 1 2
3
14 1
1 1 1
2
3
1 1 1 1
2
3
1
2
2 2 2 3
1
2 2 2 2
3
1
2 2 2 2
3
1
2
3 3 3 3
1
3 3 3 2
3
1
3 3 3 2
3
GNR401 Dr. A. Bhattacharya
 Disk File Size of the image
15

 Rows x Cols x Bands x Bytes per pixel

 For the SPOT window,
 500 x 500 x 3 x 1 = 750000 bytes ~ 750 KB
 In case of Ikonos image, storage is 2 bytes per pixel, 4
metres resolution, 4 bands
 10 km x 10 km Ikonos multispectral image size on disk =
10000/4 x 10000/4 x 4 x 2
 = 10000 x 5000 bytes ~ 50 MB
 Size of panchromatic image =
 10000 x 10000 x 2 = 10000 x 20000 bytes ~200 MB
 NOTE THE DIFFERENCE IN SIZE OF DATA!

GNR401 Dr. A. Bhattacharya

 Spectral bands and Spatial Resolution
16

 Spatial resolution is highest for panchromatic images

 Lower for multispectral images

Reason?
 In case of multispectral sensors, received energy is divided

into band-wise slices; hence lesser amount of energy to

detectors
 Compensated by increasing time of observing ground

features – hence lower spatial resolution

GNR401 Dr. A. Bhattacharya

 Image Sensing and Acquisition
17

GNR401 Dr. A. Bhattacharya

 Image Formation Model
18

GNR401 Dr. A. Bhattacharya

 Image Sampling & Quantization
19

GNR401 Dr. A. Bhattacharya

 Image Sampling & Quantization
20

GNR401 Dr. A. Bhattacharya

 Image Sampling & Quantization
21

 Sampling:
 Digitizing the coordinate values (spatial resolution)
 Quantization:
 Digitizing the amplitude values (intensity levels)

GNR401 Dr. A. Bhattacharya

 Image Quantization
22

GNR401 Dr. A. Bhattacharya

 Image Sampling
23

GNR401 Dr. A. Bhattacharya

 Image Sampling
24

GNR401 Dr. A. Bhattacharya

 Image Sampling
25

GNR401 Dr. A. Bhattacharya

 Image Sampling
26

GNR401 Dr. A. Bhattacharya

 Image Sampling 1
27

Original 256 x256 128 x 128

GNR401 Dr. A. Bhattacharya
 Image Sampling 2
28

Original 256 x256 64 x 64

GNR401 Dr. A. Bhattacharya
 Image Sampling 3
29

Original 256 x256 32 x 32

GNR401 Dr. A. Bhattacharya
 Digital Image Representation
30

GNR401 Dr. A. Bhattacharya

 Downsampling
31

GNR401 Dr. A. Bhattacharya

 Downsampling
32

GNR401 Dr. A. Bhattacharya

 Re-Sampling
33

GNR401 Dr. A. Bhattacharya

 Grey Level Quantization
34

GNR401 Dr. A. Bhattacharya

 Grey Level Quantization 1
35

Original 256 64
GNR401 Dr. A. Bhattacharya
 Grey Level Quantization 2
36

Original 256 16
GNR401 Dr. A. Bhattacharya
 Grey Level Quantization 3
37

Original 256 4
GNR401 Dr. A. Bhattacharya
 Grey Level Quantization 3
38

Original 256 2
GNR401 Dr. A. Bhattacharya
 Digital Image Representation
39

GNR401 Dr. A. Bhattacharya

 Digital Image Representation
40

GNR401 Dr. A. Bhattacharya

 Basic relationships between pixels
41

GNR401 Dr. A. Bhattacharya

 Basic relationships between pixels
42

GNR401 Dr. A. Bhattacharya

 Basic relationships between pixels
43

GNR401 Dr. A. Bhattacharya

 Basic relationships between pixels
44

GNR401 Dr. A. Bhattacharya

 Basic relationships between pixels
45

GNR401 Dr. A. Bhattacharya

 Set Logic Operations
46

GNR401 Dr. A. Bhattacharya

 Distance Function
47

GNR401 Dr. A. Bhattacharya

 Distance Function
48

GNR401 Dr. A. Bhattacharya

 Distance Function
49

GNR401 Dr. A. Bhattacharya

 Distance Function Examples
50

GNR401 Dr. A. Bhattacharya