Topic 6: OBIA

Question 1

OBIA stand for

Answer 1

Object based image analysis

Question 2

what does OBIA do

Answer 2

divides remote sensing imagery into meaningful image objects and assigning characteristics

Question 3

what is Scale in remote sensing

Answer 3

Grain
- spatial resolution
Extent
- swath width
this is a trade off

Question 4

what is the optimal scale

Answer 4

there is no single optimal scale for defining geographic entities

Question 5

why is scale an issue

Answer 5

results and conclusions may only be a result of one specific scale

Question 6

MAUP stands for

Answer 6

Modifiable areal unit problem

Question 7

what is MAUP

Answer 7

is the sensitivity of analytical results to the definition of data collection units.

Question 8

what are the two components of the MAUP

Answer 8

1. scale problem
2. aggregation problem

Question 9

what is the Scale problem in the MAUP

Answer 9

variation in results when areal units are aggregated into fewer and larger units for analysis.
eg. resampling to larger pixel sizes

Question 10

what is the aggregation problem in the MAUP

Answer 10

variation in results generated bu alternate zoning schemes at the same resolution.

Question 11

Whats the relationship between between variance and correlation coefficients

Answer 11

lower variance (smoothing) results in stronger correlation

Question 12

what are solutions to MAUP

Answer 12

1. Abandon traditional statistics
2. conduct a sensitivity analysis
3. derive an optimal spatial resolution
4. identify basic entities or objects

Question 13

what is an optimal spatial resolution

Answer 13

no simple solution, application dependent

Question 14

what is spectral classification

Answer 14

spectral classes are converted into information classes.

Question 15

what are the three types of classifications

Answer 15

1. supervised
2. unsupervised
3. hybrid

Question 16

what is Pixel mixing

Answer 16

one one pixel represents multiple classes with no one clear signal
- composite signatures

Question 17

what are competing effects of increased spatial resolution?

Answer 17

• less pixel mixing with high spatial resolution
• greater within-class variability.

Question 18

summarize the MAUP problem

Answer 18

whenever a grid is assigned to a physical area there is inherent error in the way the data is aggregated.

Question 19

OBIA as a solution for MAUP

Answer 19

provides soltion as features are not only pixel values based

Question 20

what are objects

Answer 20

are basic entities within an image that are composed of h-res pixel groups
- we use shape, texture, colour and context to define shapes

Question 21

intigration of Image objects to GIS

Answer 21

pixels: raster data model
objects: vector data model

Question 22

OBIA provide what aspects for analysis

Answer 22

Spectral Variables
Shape variables
Texture Variables
Context

Question 23

what are spectral variables

Answer 23

feautres related to the value of pixels within an object

Question 24

what are Shape variables

Answer 24

Features to the shape of an image object, length, width, direction, area

Question 25

texture variables

Answer 25

features related to spatial patterns of pixels within an object

Question 26

what is Context in a OBIA

Answer 26

Features related to the spatial setting of an object within scene and/or its relationship to other objects in the scene

Question 27

what does one-to-one correspondance in OBIA mean

Answer 27

the geographic entity is perfectly selected and the correct spatial scale is being used.

Question 28

what is segmentation

Answer 28

pulling discrete regions out of an image

Question 29

how is a segmentation preformed

Answer 29

- using premade algorithms - expert knowledge of your image and application - calibrate, trial and error

Question 30

what are the two types of segmentation algorthims

Answer 30

1 discontinuity based 2 similarity based

Question 31

what is a discontinuity based algorithm

Answer 31

image i sbroken based on abrubt changes in intensity, edge detection

Question 32

what is similarity based algorithm

Answer 32

image is broken using regions based on pre-defined criteria using spectral or spatial information

Question 33

Laplacian edge detection

Answer 33

-discontinuity based Segmentation algo -directionally invariant edge detection

Question 34

Sobel edge detection

Answer 34

another algo for dicontinuity based segmentation

Question 35

thresholding

Answer 35

similarity based segmentation. partioning a histogram

Question 36

region based segmentation

Answer 36

similarity based segmentaion that partitions an image based on region growing approach - seed points - growing regions

Question 37

what is Trimble

Answer 37

a segmentation algorithm that is a region merging technique - process can be repeated at many scales

Question 38

OBIA workflow

Answer 38

depends on purpose

Question 39

partitioning an image into meaningful image objects requires....

Answer 39

an effective segmentation technique

Question 40

melt ponds

Answer 40

ponds that form on the top of glaciers that increase melting

Question 41

is C-band frequency SAR Capable of providing a seasonal prediction of spring melt ponds?