Lec 10 - Information Extraction

Question 1

Purposes of Image Extraction (4)

Answer 1

(1) Categorizing data
(2) Data simplification
(3) Data interpretation
(4) Mapping

Question 2

Overall objective of classification

Answer 2

Automatically categorize all pixels in an image into land cover classes or themes

Question 3

Selection of Classification (3)

Answer 3

(1) The nature of the data being analyzed
(2) The computational resources available
(3) The intended application of the classified data

Question 4

Pattern Recognitions (3)

Answer 4

(1) Spatial pattern recognition
(2) Spectral pattern recognition
(3) Temporal pattern recognition

Question 5

Use spatial context to distinguish between different classes

Answer 5

Spatial Pattern Recognition

Question 6

Most widely-used pattern recognition that distinguish between different land cover classes from differences in the spectral reflectance

Answer 6

Spectral Pattern Recognition

Question 7

The ability to distinguish patterns based on spectral or spatial considerations that may vary over the year

Answer 7

Temporal Pattern Recognition

Question 8

Land Cover Mapping and Applications of Remote Sensing

Answer 8

(1) Understanding of the type and amount of land cover in an area is an important characteristic
(2) Remote sensing has become a powerful tool for land cover identification and classification
(3) The investment in the development of this technology has contributed to Precision Agriculture

Question 9

Steps in Thematic Information Extraction from Satellite Images

Answer 9

(1) Definition of the mapping approach
(2) Geographical stratification
(3) Image segmentation (for object-oriented classification)
(4) Feature Identification and Selection
(5) Classification

Question 10

The unit to which the classification algorithms will be applied.

Answer 10

Spatial unit of analysis

Question 11

The study area is divided into smaller areas (strata) so that each strata can be processed independently

Answer 11

Geographical stratification

Question 12

The division of an image into spatially continuous, disjoint and homogenous regions

Answer 12

Image segmentation

Question 13

The manipulation and selection of features are used to reduce the number of features without sacrificing accuracy

Answer 13

Feature Identification and Selection

Question 14

The process of partitioning an image data set into a discrete number of classes in accordance with specific criteria that are based, in part, on the individual image point data values (recognize patterns).

Answer 14

Classification

Question 15

This depends on the type of information you want to extract from the original data. May simply represent areas that look different to the computer or may be associated with known features on the ground.

Answer 15

Classes

Question 16

Types of Classification (based on type of learning

Answer 16

(1) Supervised
(2) Unsupervised
(3) Hybrid

Question 17

Type of classification that requires “training pixels”, pixels where both the spectral values and the class is known.

Answer 17

Supervised

Question 18

Type of classification where no extraneous data is used: classes are determined purely on difference in spectral values.

Answer 18

Unsupervised

Question 19

Use unsupervised and supervised classification together

Answer 19

Hybrid

Question 20

Supervised Classification Steps

Answer 20

(1) Training stage
(2) Classification stage
(3) Output stage

Question 21

Pixel observations from training sites are plotted on scatter diagrams and are used as a basis of setting up the statistical parameters used to classify pixels outside these sites.

Answer 21

Classification stage

Question 22

This type of classifier determines the mean value (DN) of each class in each band, and then assigns unknown pixels to classes whose means are most similar to the value of the unknown pixel.

Answer 22

Minimum Distance to Means

Question 23

Benefit of Minimum Distance to Means

Answer 23

This method is quite computationally efficient since it is mathematically simple, which made it a good choice before the advent of modern computers.

Question 24

Drawback of Minimum Distance to Means

Answer 24

Insensitive to different degrees of variance in spectral response data

Question 25

The most powerful classifier in common use. A statistical decision rule that examines the probability function of a pixel for each of the classes, and assigns the pixel to the class with the highest probability.

Answer 25

Maximum Likelihood

Question 26

Benefit of Maximum Likelihood

Answer 26

Takes variation in spectral response into consideration

Question 27

Drawback of Maximum Likelihood

Answer 27

Computationally intensive; multimodal or non-normally distributed classes require extra care when training the classifier, if high accuracy is to be achieved

Question 28

This works by delineating the boundaries of a training class using straight lines.

Answer 28

Parallelepiped

Question 29

Benefit of Parallelepiped

Answer 29

This method is computationally fast, simple to train and use

Question 30

Drawbacks of Parallelepiped

Answer 30

Using straight lines to delineate the classes limits the method’s effectiveness. Also, having pixels classified as unknown may be undesirable for some applications

Question 31

It is based on correlations (statistics) between variables by which different patterns can be identified and analyzed.

Answer 31

Mahalanobis Distance

Question 32

Any individual pixel is compared to each discrete cluster to see which one it is closest to.

Answer 32

Unsupervised Classification

Question 33

Advantage of Unsupervised Classification

Answer 33

No extensive prior knowledge required opportunities of human error minimized unique classes recognized as distinct units

Question 34

Disadvantage of Unsupervised Classification

Answer 34

Limited control over classes and identities

Question 35

A more sophisticated version of the K-Means classifier which allows classes to be created and destroyed.

Answer 35

Iterative self-organizing data analysis (ISODATA)

Question 36

Each point has a degree of belonging to clusters (fuzzy logic) rather than belonging completely to just one cluster.

Answer 36

Fuzzy C Means