Classification techniques

There are two basic approaches to the classification process: supervised and unsupervised classification. With supervised classification, one provides a statistical description of the manner in which expected land cover classes should appear in the imagery, and then a procedure (known as a classifier) is used to evaluate the likelihood that each pixel belongs to one of these classes. With unsupervised classification, a very different approach is used. Here another type of classifier is used to uncover commonly occurring and distinctive reflectance patterns in the imagery, on the assumption that these represent major land cover classes. The analyst then determines the identity of each class by a combination of experience and ground truth (i. e., visiting the study area and observing the actual cover types) (Eastman 2003). Three essential parts are vital in a LUC mapping in classification stage; training, classifying and testing (accuracy assessment).

4.1.1 Classifiers

In this chapter four different classifiers and approaches were evaluated in the example of Landsat TM sub-scenes recorded over Eastern Mediterranean coastal part. Methods and performances were assessed based on accuracy, capability and applicability. This assessment covered traditional (minimum distance, maximum likelihood, linear discriminant analyses), machine learning (decision tree, artificial neural network, support vector machine), fuzzy (linear mixture modeling, fuzzy c-means, artificial neural network, regression tree) and object based classifiers for LUC mapping. The summary of the techniques and classifiers for various purposes were provided in table 4.

Criteria

Categories

Characteristics

Example of classifiers

Whether

Supervised

Land cover classes are defined.

Maximum likelihood

training

Classification

Sufficient reference data are available

(MLC), minimum

samples are

approaches

and used as training samples. The

distance (MD),

used or no

signatures generated from the training samples are then used to train the classifier to classify the spectral data into a thematic map.

Artificial neural network (ANN), decision tree (DT) classifier.

Unsupervised

Clustering-based algorithms are used

ISODATA, K-means

classification

approaches

to partition the spectral image into a number of spectral classes based on the statistical information inherent in the image. No prior definitions of the classes are used. The analyst is responsible for labeling and merging the spectral classes into meaningful classes.

clustering algorithm.

Whether

Parametric

Gaussian distribution is assumed. The

MLC and Linear

parameters

classifiers

parameters (e. g. mean vector and

discriminant analysis

such as mean vector and covariance matrix are used or not

covariance matrix) are often generated from training samples. When landscape is complex, parametric classifiers often produce ‘noisy’ results. Another major drawback is that it is difficult to integrate ancillary data, spatial and contextual attributes, and non-statistical information into a classification procedure.

(LDA)

Non-

No assumption about the data is

ANN, DT, Support

parametric

required. Non-parametric classifiers do

vector machine

classifiers

not employ statistical parameters to calculate class separation and are especially suitable for incorporation of non-remote-sensing data into a classification procedure.

(SVM), evidential reasoning, expert system.

Which kind of

Per-pixel

Traditional classifiers typically develop

MLC, MD, SVM,

pixel

information is used

classifiers

a signature by combining the spectra of all training-set pixels from a given feature. The resulting signature contains the contributions of all materials present in the training-set pixels, ignoring the mixed pixel problems.

ANN, DT

Subpixel

The spectral value of each pixel is

Fuzzy-set classifiers,

classifiers

assumed to be a linear or non-linear combination of defined pure materials (or endmembers), providing proportional membership of each pixel to each endmember.

subpixel classifier, spectral mixture analysis.

Criteria

Categories

Characteristics

Example of classifiers

Which kind of

Object-

Image segmentation merges pixels into

eCognition.

pixel

oriented

objects and classification is conducted

information is used

classifiers

based on the objects, instead of an individual pixel. No GIS vector data are used.

Per-field

GIS plays an important role in per-field

GIS-based

classifiers

classification, integrating raster and vector data in a classification. The vector data are often used to subdivide an image into parcels, and classification is based on the parcels, avoiding the spectral variation inherent in the same class.

classification

approaches.

Whether

Hard

Making a definitive decision about the

MLC, MD, ANN, DT,

output is a

definitive

decision about land cover class or not

classification

land cover class that each pixel is allocated to a single class. The area estimation by hard classification may produce large errors, especially from coarse spatial resolution data due to the mixed pixel problem.

SVM

Soft (fuzzy)

Providing for each pixel a measure of

Fuzzy-set classifiers,

classification

the degree of similarity for every class. Soft classification provides more information and potentially a more accurate result, especially for coarse spatial resolution data classification.

subpixel classifier, spectral mixture analysis.

Whether

Spectral

Pure spectral information is used in

Maximum likelihood,

spatial

classifiers

image classification. A ‘noisy’

minimum distance,

information is

classification result is often produced

artificial neural

used or not

due to the high variation in the spatial distribution of the same class.

network.

Contextual

The spatially neighbouring pixel

Iterated conditional

classifiers

information is used in image classification

modes, point-to – point contextual correction, and frequency-based contextual classifier.

Spectral-

Spectral and spatial information is

ECHO, combination

contextual

used in classification. Parametric or

of para metric or

classifiers

non-parametric classifiers are used to generate initial classification images and then contextual classifiers are implemented in the classified images.

non-parametric and

contextual

algorithms.

Table 4. A taxonomy of image classification methods (Lu and Weng 2007).